From mild cognitive impairment to moderate Alzheimer's disease: Associations between decision-making capacity and awareness

Abstract

Background

Evidence on how distinct stages of mild cognitive impairment (MCI) and Alzheimer's disease (AD) affect decisional capacity and disease awareness remains limited.

Objective

We aim to examine the impairment of decision-making capacity, hypothesizing that people with MCI would demonstrate greater capacity and awareness than those with AD. We also aim to explore associations with cognitive and clinical variables.

Methods

We conducted a cross-sectional study including 137 participants: MCI (n = 37), mild AD (n = 50), and moderate AD (n = 50). Standardized assessments were used to evaluate domains of decision-making capacity and awareness of the illness.

Results

People with MCI performed significantly better in understanding, appreciation, and reasoning domains, while expression of choice was preserved across all groups. Disease awareness was higher in MCI, whereas AD participants showed marked deficits, particularly in cognitive and functional domains. Differences in socioemotional awareness were less consistent.

Conclusions

Our findings indicate a decline, from MCI to moderate AD, in decisional capacity and awareness, with expression of choice remaining relatively intact. These results have important ethical and clinical implications for shared decision-making and dementia care.

Keywords

Alzheimer's disease awareness decision-making capacity impairment in cognitive and clinical domains mild cognitive impairment

Introduction

Decision-making capacity refers to the ability to make informed and autonomous choices. Such capacity is not a static trait but is both time- and task-dependent, meaning that an individual's ability to make informed choices may fluctuate depending on the specific decision at hand and the moment involved.^1,2 The process involves identifying a decision, gathering relevant information, evaluating alternative outcomes, and selecting an appropriate action from several possible options.¹ It is a complex process that engages multiple cognitive mechanisms to determine the proper course of action, including goal-directed motivation, weighing the likely consequences of different options, and deciding which expected consequences best fit the objectives.²

Decision-making capacity encompasses four key domains^3–5: (a) understanding the important information involves gathering, storing, and recalling relevant information; (b) appreciating the implications of contextualized information is the capacity related to personal values concerning the benefits and risks of a choice; (c) reasoning about the information is the logical process of comparing the alternative answer and indicating the reason for the choice, which requires logically consistent reasoning; and (d) expressing a choice is the ability to change a choice and maintain its consistency until its implementation. Deficits in any of these domains impair the individual's decision-making capacity.

Decision-making capacity is often described as declining with age. The relationship is more complex and variable than simply an age-related trajectory.² Normal aging can moderately affect decision-making capacity, but these changes are not inevitable and may differ substantially between individuals and types of decisions. Impairments are more pronounced in pathological aging, particularly in neurodegenerative disorders such as Alzheimeŕs disease (AD). Recent studies^6–8 indicate that people with AD present deficits in their decision-making capacity, especially understanding and reasoning domains. Furthermore, a meta-analysis by Parmigiani found that compared to controls, people with mild cognitive impairment (MCI) scored lower in the domains of comprehension, appreciation, and reasoning. Studies indicate that individuals with AD often demonstrate relatively better performance in comprehension compared to their deficits in reasoning and appreciation, though overall capacity remains impaired..^6–8 In contrast, meta analytic evidence shows that people with MCI tend to score lower across the comprehension, appreciation, and reasoning domains, reflecting a broader but less severe impairment than in AD.⁷ Importantly, while MCI is associated with higher risk of impaired decision-making capacity compared to healthy controls, this risk is still lower than in AD. Beyond cognitive domains, clinical variables such as functional status and neuropsychiatric symptoms (apathy, depression, or behavioral disturbances) have also been shown to significantly affect decision-making capacity, underscoring that impairment is not solely determined by age or diagnosis but by a constellation of cognitive and clinical variables.⁷

Other studies^9,10 also demonstrate that decision-making capacity is compromised in AD and MCI, but with greater impairment in AD. Furthermore, reduced decision-making capacity is closely related to lower executive functions and memory.

Considering decision-making in neurocognitive diagnoses, it is important to understand the individual's awareness of their illness and associated deficits. Awareness is a multidimensional construct that encompasses several domains such as cognitive functioning, relationships, or functional impairments.^11–13 Awareness is the capacity to recognize changes caused by onés illness, and it operates on biological, social, and psychological levels. It should not be seen merely as a symptom of disease, as the pattern of impairment varies depending on the specific domain assessed.¹²

A meta-analysis showed that individuals with MCI are typically aware of their neuropsychological deficits. However, their level of awareness can vary depending on cognitive status, language skills, and memory abilities.¹⁴ Other studies indicate that awareness did not significantly differ between individuals with MCI and those with mild AD.⁸ In contrast, people with AD displayed impaired recognition of deficits across all awareness domains. Specifically, people with AD were less aware of their deficits compared to those with MCI and often did not attribute their symptoms to the illness.^15–17 Research shows that in mild AD and MCI, awareness of deficits is more strongly tied to affective (mood and emotional regulation) and behavioral changes (motivation, social appropriateness or impulse control) than to purely cognitive decline.¹⁶ However, there is a growing consensus that unawareness becomes more frequent as dementia progresses.¹⁸

In a recent study,⁸ we examined the relationship between decision-making capacity and awareness in individuals with AD. Our findings indicated that those who were aware of their cognitive functioning and health status were more likely to be deemed competent in their decision-making abilities. Furthermore, we explored how decision-making capacity and awareness varied according to age at onset of dementia. Our results showed that individuals with young-onset AD (YOAD) were more cognitively impaired but exhibited greater awareness of their cognitive deficits and overall health condition. This suggests that better awareness may lead to more accurate understanding of their situation within the YOAD group.^8,19

While previous studies^8,19 have explored the relationship between awareness and decision-making capacity in individuals with AD, no study to date has systematically examined these constructs across the continuum from MCI to AD. This study aims to fill that gap. In the context of cognitive impairment and dementia, the ability to understand, appreciate, reason about, and express choices and their relationship with awareness has profound ethical and clinical implications. Preserving decisional capacity is central to shared decision-making and to safeguarding patients’ rights in dementia care. In this study, we aim to analyze the impairment of decision-making capacity among people with MCI and AD. We hypothesize that people with MCI have significantly higher decision-making capacity and better awareness when compared to those with AD. We also aim to investigate the relationship between decision-making capacity, cognition and clinical (severity of disease, functionality, neuropsychiatric symptom and mood) impairment in people with MCI and AD.

Methods

This cross-sectional study followed the STROBE guidelines for observational research.

Participants

Sample size was calculated with the G*Power software.²⁰ For a one-way ANOVA with three groups, assuming a moderate effect size (f = 0.27), level of significance α=0.05, and statistical power 0.80, the minimum required sample was 135 participants. Participants (n = 137) were allocated into three groups based on the Clinical Dementia Rating (CDR) scale: MCI (CDR=0.5, n = 37), mild AD (CDR=1, n = 50), and moderate AD (CDR=2, n = 50).

All participants had been diagnosed according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5)²¹ and were attending the Center of Alzheimeŕs Disease and Related Disorders, Institute of Psychiatry, Universidade Federal do Rio de Janeiro, Brazil. A psychiatrist performed the diagnosis using clinical interviews, cognitive screening tests, laboratory tests, and imaging analyses.

Only individuals with mild and moderate AD and MCI according to the CDR^22,23 were included. We excluded people with a history of psychiatric or other neurological disorders, aphasia, epilepsy, head trauma, or alcohol and drug abuse.

The primary family caregiver was defined as the person who was mostly responsible for care of the person with AD or MCI. All caregivers were previously informed of the diagnosis by the psychiatrist.

Ethics

The study was approved by the Institutional Review Board of the Institute of Psychiatry (IPUB), Universidade Federal do Rio de Janeiro (UFRJ). All participants signed the informed consent form after receiving a detailed explanation of the study and use of the resulting data.

Procedures

The assessments were conducted by trained psychologists from the research team. Individuals with AD and MCI were interviewed separately from their caregivers. For interviews with participants with AD and MCI, the research assistant read the questions, accompanied by a display in large font of the choices of answers. In case of confusion involving the question or choices of answers, the research assistant attempted to clarify the participant's doubts.

Participants with AD or MCI were assessed for their decision-making capacity, awareness of their condition, and cognitive functioning. Caregivers provided information about the individuals with AD or MCI, including demographic details, their ability to perform activities of daily living, neuropsychiatric symptoms, and mood. The instruments were administered in the same order for all participants.

Instruments

Decision-making capacity. The MacArthur Competence Assessment Tool for Treatment (MacCAT-T) is a semi-structured interview designed to evaluate an individual's capacity to consent to treatment. The instrument focuses on four key decision-making abilities: understanding, appreciation, reasoning, and expression of choice. The “Understanding” section assesses the individual's grasp of their illness, treatment options, and the associated risks and benefits. Here, the interviewee was asked to paraphrase the information provided to demonstrate their comprehension. The “Appreciation” section evaluates whether the individual can apply the information to their own circumstances (appreciation of the illness) and recognize the potential benefits of the proposed treatment (appreciation of the treatment). The “Reasoning” section asks the person to discuss the consequences of various treatment alternatives (consequential reasoning), compare different options (comparative reasoning), and identify potential outcomes that were not mentioned by the interviewer (generating consequences). This section also assesses the logical consistency of the individual's choice. Finally, in the “Expression of Choice” section, the individual must indicate a clear preference for one treatment over others. Each item is scored as follows: 2 (adequate), 1 (partially adequate), and 0 (inadequate). There are specific quantitative scores for each ability: 0–6 for understanding, 0–4 for appreciation, 0–8 for reasoning, and 0–2 for expression of a choice. These normative patterns were used to contextualize the scores in the present study. It is important to note that there is no overall score for the MacCAT-T, nor is there a cutoff score used to determine whether an individual is competent to make decisions about their treatment. Instead, the interviewer's clinical judgment must consider all relevant information.^24,25

Awareness of the disease. The Assessment Scale of Psychosocial Impact of the Diagnosis of Dementia (ASPIDD) is a 30-question tool that gathers information based on reports from both individuals with AD or MCI and their caregivers. It evaluates the participants’ awareness of their disease across several domains: awareness of cognitive functioning and health condition, functional activity impairments, emotional state, social functioning, and relationships. The awareness score is the degree of discrepancy between responses by the person with AD or MCI and their caregiver, with one point scored for each discrepant response. The total scores for awareness range from preserved (0–4), mildly impaired (5–11), and moderately impaired (12–17) to absent (over 18). The scores for domains are given as: awareness of cognitive functioning and health condition (0–7; >1 impaired), emotional state (0–5; >1 impaired), social functioning and relationship (0–7; >1 impaired), and functional activity impairment (0–11; >1 impaired).¹²

Dementia severity

The CDR measures the severity of the dementia. The stages range from 0 (no dementia) to 3 (severe dementia) according to the degree of impairment in cognitive and behavioral aspects and activities of daily living (ADL).^22,23

Cognition. The Mini-Mental State Examination (MMSE) includes tasks of orientation, registration, short-term memory, language use, comprehension, and basic motor skills. The total score ranges from 0 to 30. Lower scores indicate more impaired cognition.^26,27

Functionality. The Pfeffer Functional Activities Questionnaire (PFAQ) is a caregiver-reported inventory that evaluates activities of daily living. Scores for each item range from normal (0) to dependent (3), with a total of 30 points. Higher scores indicate worse functional status.^28,29

Neuropsychiatric symptoms. Neuropsychiatric Inventory (NPI) is a questionnaire and evaluates the presence of delusions, hallucinations, dysphoria, anxiety, agitation/aggression, euphoria, disinhibition, irritability/lability, apathy, aberrant motor activity, night-time behavior disturbances, and appetite and eating abnormalities. Each item is rated for frequency and intensity. The total score ranges from 0 to 144 points. Higher scores indicate greater presence of neuropsychiatric symptoms.^30,31

Depressive symptoms. The Cornell Scale for Depression in Dementia (CORNELL) assesses mood symptoms, physical symptoms, circadian functions, and behavioral symptoms related to depression. Scores above 13 indicate presence of depression.^32,33

Statistical analysis

Statistical analyses were conducted using SPSS for Windows, version 24.0. The Kolmogorov-Smirnov test was employed to assess normal distribution of the data, while Levenés test was used to verify normal distribution and homoscedasticity, respectively.

The first step involved analysis of the participants’ sociodemographic and clinical data. We used the independent t-test, chi-square test, and analysis of variance (ANOVA) to evaluate the relationship between sociodemographic and clinical variables in both groups.

For effect size measurement, Cohen's d was applied (small: 0.520; medium: 0.50; large: 0.80), besides Phi (ranging from −1 to 1, with higher values indicating a stronger correlation). Additionally, for the chi-square test, we used Crameŕs V, where 0 indicates complete independence and 0.1 indicates complete dependence or association regarding differences between groups in sociodemographic and clinical characteristics.

The WRS2 package uses bootstrap methods and robust techniques, including Welch´s correction, to evaluate differences between groups, making it particularly useful in contexts where data may contain outliers or non-normal distributions. The bootstrap method in this context makes the analysis less sensitive to deviations from normality and heteroscedasticity, as the technique resamples the data and provides more robust variance estimates. The robust ANOVA analysis was made using the WRS2 package of the R Project for Statistical Computing (2024). This package offers an alternative to the situations of traditional assumptions in ANOVA, like not displaying normality and homogeneity.

Robust ANOVA used a p-value of <0.05, indicating a significant difference between the tested groups. In the post hoc test, pairwise group comparisons were conducted using the Dwass-Stell-Critchlow-Fligner procedure, which provides nonparametric multiple comparisons among independent groups, similarly used a p-value of <0.05, indicating a significant difference between the groups. Violations of the assumptions of normality and homogeneity of variances were detected; therefore, nonparametric difference tests (e.g., Kruskal–Wallis) were employed. In accordance with this approach, post hoc analyses were conducted using nonparametric procedures. Specifically, the Dwass–Steel–Critchlow–Fligner post hoc multiple comparisons test was applied. This test is based on pairwise median comparisons utilizing rank-based statistics, thereby providing a rigorous framework for identifying specific intergroup differences under nonparametric conditions.

To investigate the explanatory relationship between decision-making capacity domains and awareness and the clinical variables, we performed multiple linear regression analyses (stepwise method) for each of the dependent variables according to CDR level.

For all analyses, α level was set at p ≤ 0.05.

Results

Sociodemographic and clinical characteristics of people with MCI and mild and moderate AD

Individuals with MCI (N = 37) had a mean age of 72.78 years (SD = 6.90), with 3.40 years’ (SD = 3.13) duration of the illness. People with mild AD (N = 50) had a mean age of 77.58 years (SD = 6.34), with 4.88 years’ (SD = 2.97) duration of the illness. In the group of people with moderate AD (N = 50), mean age was 74.27 years (SD = 10.52), with 5.05 years’ (SD = 3.37) duration of the disease. Most individuals in the MCI group had more years of education (11.22 years; SD = 4.18), when compared to the groups with mild AD (7.40; SD = 3.74) and moderate AD (7.92; SD = 4.05). There was a significant difference in age (p=<0.05; d = 0.61) with moderate effect size and disease duration (p=<0.05; 0.22) and years of schooling (p = 0.001; d = 0.15), although the effect sizes were small.

Participants with moderate AD showed significantly greater cognitive impairment, with a large effect size, when compared to people with MCI and mild AD (p = 0.001; d = 0.87).

Functionality showed significant differences between groups (p=<0.001). People with MCI displayed better functionality (5.13; SD = 6.67). The effect size was large (0.77), indicating a strong effect between groups. Neuropsychiatric symptoms also differed significantly between groups (p=<0.001), increasing from people with MCI (11.11; SD = 10.13) and reaching the highest level in people with moderate AD (22.20; SD = 16.36). In contrast, depressive symptoms did not differ significantly between groups (p = 0.12), with similar scores between groups and small effect size (0.25).

Table 1 shows the comparison of demographics and clinical data between people with MCI, mild AD, and moderate AD.

Table 1.

Sociodemographic and clinical data.

Sociodemographic and clinical data
Variables	MCI (N = 37)	Mild AD (N = 50)	Moderate AD (N = 50)	Significance	Effect size
Females N (%)	31 (83.8)	30 (60)	36 (72)	0.05^b	0.19
Males N (%)	6 (16.2)	20 (40)	24 (28)	0.05^b	0.76
Age (mean—SD)	72.78 (6.90)	77.58 (6.34)	74.27 (10.52)	0.05^a	0.61
Age of on-set (mean—SD)	69.40 (6,89)	72.70 (6.95)	69.28 (10.32)	0.08	0.72
Disease duration (mean—SD)	3.40 (3.13)	4.88 (2.97)	5.05 (3.37)	0.05^b	0.22
Schooling (mean—SD)	11.22 (4.18)	7.40 (3.74)	7.92 (4.05)	<0.001^a	0.15
CDR (%)	0.5 (27.01)	1 (36.50)	2 (36.50)	<0.001^a,c,d	0.80
MMSE (mean—SD)	25.35 (3.46)	21.32 (3.18)	14.68 (2.73)	<0.001^a,c,d	0.87
PFAQ (mean—SD)	5.13 (6.67)	14.16 (7.93)	22.70 (6.54)	<0.001^a,c,d	0.77
NPI (mean—SD)	11.11 (10.13)	13.46 (11.17)	22.20 (16.36)	<0.001^a,c,d	0.42
CORNELL (mean—SD)	7.97 (6.30)	7.70 (5.53)	9.85 (5.89)	0.12	0.25

CDR: Clinical Dementia Rating; MMSE: Mini-Mental State Examination; PFAQ: Pfeffer Functional Activities Questionnaire; NPI: Neuropsychiatric Inventory; CORNELL: The Cornell Scale for Depression in Dementia. Effect size Cohen d: p ≤ 0.05.

Significance: ^aKruskal-Wallis test, mean (standard deviation); ^b Chi-square test, n (%); p < 0.05.

Pairwise comparison analysis:

Significant difference between MCI and Mild AD.

Significant difference between MCI and Moderate AD.

Significant difference between Mild AD and Moderate AD.

Differences between awareness and decision-making capacity

Regarding awareness and its domains, there were no significant differences between groups in the awareness of social functioning and relationships (p = 0.36) or in the awareness of emotional state domains (p = 0.32). People with MCI were more aware of their functional activity impairment (p=<0.001; d = 1.43) and of their cognitive deficits and health condition (p=<0.001; d = 0.86) and total ASPIDD (p=<0.001; d = 0.99), with large effect size between groups. The group with mild AD was more aware than the group with MCI in the social functioning and relationship domain (p = 0.36; d = 0.17), with small effect size. The group with moderate AD had the worst awareness compared to the group of people with MCI and mild AD, as expected.

There were significant differences between the domains of decision-making capacity, with moderate effect size between groups in all the domains except for the expression of choice domain (p = 0.25; d = 0.19). People with MCI had better understanding (p=<0.001; d = 0.72), appreciation (p=<0.001; d = 0.61), and reasoning (p=<0.001; d = 0.77). People with mild AD showed better expression of choice (p = 0.27; d = 0.19) compared to people with MCI. To assist interpretation, these values fall within the ranges typically observed in normative samples of older adults and people with AD as reported in the literature.

Table 2 summarizes the differences between awareness and the domains of decision-making capacity.

Table 2.

Decision-making capacity domains and awareness.

Decision-Making capacity domains and awareness
Variables (mean—SD)	MCI (N = 37)	Mild AD (N = 50)	Moderate AD (N = 50)	Significance	Effect size
MacCAT-T Understanding	5.15 (1.33)	3.84 (1.41)	3.21 (1.31)	<0.001^c,d,e	0.72
MacCAT-T Appreciation	3.60 (0.80)	2.92 (1.17)	2.27 (1.22)	<0.001^d,e	0.61
MacCAT-T Reasoning	6.40 (2.29)	3.28 (1.68)	2.46 (1.83)	<0.001^c,d,e	0.77
MacCAT-T Expression of a choice	1.62 (0.55)	1.72 (0.67)	1.54 (0.81)	0.28	0.19
ASPIDD Total	7.07 (4.93)	8.96 (4.66)	11.88 (5.19)	<0.001	0.99
ASPIDD Cognitive Deficits and Health Conditions	1.88 (1.50)	2.41 (1.16)	3.27 (1.82)	<0.001^d	0.86
ASPIDD Emotional State	1.40 (1.27)	1.67 (1.54)	1.65 (1.52)	0.32	0.17
ASPIDD Social Functioning and Relationships	1.66 (1.49)	1.10 (1.05)	1.41 (1.17)	0.36	0.17
ASPIDD Functional Activity Impairment	2.10 (2.19)	3.75 (2.46)	5.55 (2.74)	<0.001^c,d,e	1.43

MacCAT-T: The MacArthur Competence Assessment; ASPIDD: Assessment Scale of Psychosocial Impact of the Diagnosis of Dementia; Effect size Cohen d; p ≤ 0.05.

Significance: ^aKruskal-Wallis test, mean (standard deviation); ^bChi-square test, n (%); p < 0.05.

Pairwise comparison analysis:

Significant difference between MCI and Mild AD.

Significant difference between MCI and Moderate AD.

Significant difference between Mild AD and Moderate AD.

Regression analysis

Regression models were constructed for all domains of decision-making capacity. Not all were significant, however in people with MCI, all domains were significant predictors. Specifically, the presence of neuropsychiatric symptoms significantly predicted appreciation (p < 0.001), understanding (p = 0.011) and reasoning domains (p = 0.033). The cognitive level and presence of neuropsychiatric symptoms predicted the expression of choice domain (p < 0.001).

Among people with mild AD, the presence of neuropsychiatric symptoms significantly predicted understanding, reasoning, and expression of choice (p = 0.01). Neuropsychiatric symptoms (p=<0.001) and level of functionality (p = 0.039) predicted the expression of choice domain.

In people with moderate AD, the presence of neuropsychiatric symptoms significantly predicted the domains of understanding, appreciation, and expression of choice (p < 0.001). Furthermore, total awareness predicted understanding and appreciation domains, while level of functionality predicted expression of choice (p = 0.011).

Table 3 shows the results of the regressions models.

Table 3.

Regression models of predictors of decision-making capacity according to the level of cognitive impairment.

Regression
Variables	B	t	R²	R² Adjust	β	Significance
MacCAT-T Understanding MCI			0.40	0.35		0.018
CDR	−0.99	0.44			−0.35	0.161
NPI	−0.51	0.44			−0.32	0.011
MacCAT-T Appreciation MCI			0.37	0.33		0.016
NPI	−0.34	0.54			−0.44	<0.001
MacCAT-T—Reasoning MCI			0.13	0.11		0.033
NPI	0.03	0.63			−0.21	0.033
MacCAT-T Expression of a choice MCI			0.59	0.55		0.014
MMSE	0.08	0.60			−0.22	<0.001
NPI	0.10	0.60			−0.38	<0.001
MacCAT-T Understanding Mild DA			0.30	0.27		0.015
NPI	−0.77	0.60			−0.39	0.014
MacCAT-T Appreciation Mild DA			0.27	0.22		0.071
NPI	−0.56	0.60			−0.36	<0.001
MacCAT-T Reasoning Mild DA			0.26	0.23		0.009
NPI	0.31	0.18			−0.45	0.007
MacCAT-T Expression of a choice Mild DA			0.39	0.34		0.039
NPI	−0.31	0.19			−0.35	<0.001
PFAQ	−0.03	0.09			−0.45	0.039
MacCAT-T Understanding Moderate DA			0.53	0.50		<0.001
ASPIDD Total	−0.17	0.19			−0.43	<0.001
NPI	−0.32	0.10			−0.45	<0.001
MacCAT-T Appreciation Moderate DA			0.19	0.15		0.038
NPI	−0.22	0.11			−0.43	0.028
ASPIDD Total	−0.08	0.03			−0.03	0.038
MacCAT-T Reasoning Moderate DA			0.22	0.13		0.096
ASPIDD Cognitive Functioning and Health Condition	−0.25	0.61			−0.44	0.122
PFAQ	0.10	0.03			−0.69	0.028
NPI	−0.30	0.37			−0.23	0.064
ASPIDD Social and Functioning Relationship	−0.43	0.55			−0.21	0.096
MacCAT-T Expression of a choice Moderate DA			0.32	0.28		0.011
NPI	−0.23	0.36			−0.43	0.011
PFAQ	0.04	0.40			−0.03	0.011

MacCAT-T: The MacArthur Competence Assessment; CDR: Clinical Dementia Rating; NPI: Neuropsychiatric Inventory; MMSE: Mini-Mental State Examination; PFAQ: Pfeffer Functional Activities Questionnaire; ASPIDD: Assessment Scale of Psychosocial Impact of the Diagnosis of Dementia; Effect size Cohen d; p ≤ 0.05.

Discussion

We investigated the relationship between domains of decision-making capacity and domains of awareness in people with MCI and AD. This is the first study to explore this relationship across MCI, mild AD, and moderate AD. We also aimed to analyze impaired decision-making capacity and its association with cognitive and clinical variables in these populations. The results demonstrate a consistent pattern throughout clinical worsening of AD and contribute to a deeper understanding of the factors influencing autonomy across different stages of the disease. The sociodemographic characteristics demonstrated that people with moderate AD showed significantly higher cognitive impairment, with large effect size (d = 0.87), confirming advancing cognitive decline as the disease evolves. These findings validate that the sample reflects a typical distribution along the worsening of AD.

We hypothesized that people with MCI would demonstrate significantly higher decision-making capacity and greater awareness compared to those with AD. Our results confirmed the hypothesis. We found a progressive decline in decisional capacity from MCI to moderate AD, with the MacCAT-T domains of understanding, appreciation, and reasoning most affected. In contrast, the ability to express a choice was relatively preserved across groups.

This result is consistent with previous studies, suggesting that the ability to express a choice is often preserved longer than more cognitively demanding domains such as understanding, appreciation, and reasoning.^8,19,34,35 The lack of difference in the expression of choice domain, even among people with moderate AD, can also be explained by randomness and may not be linearly related to the cognitive impairment or disease severity.³⁶ Therefore, variability in this domain may arise from random factors or individual differences unrelated to the progression of AD. The criterion focuses on how decisions are made rather than the outcome of the participant's choice, since participants have the right to make unreasonable choices. This distinction reflects the ethical principle that participants retain the right to make decisions that may appear unreasonable, provided the decision-making process itself is intact. The fact that individuals with MCI and mild AD still performed relatively well in several of these domains reinforces the importance of not presuming incapacity based only on diagnosis. Instead, structured and individualized assessments are essential to ensure accurate determination of decision-making ability and appropriate support when needed.

Regarding awareness, MCI participants exhibited greater global awareness compared to both AD groups. The decline in awareness was particularly evident in the cognitive and functional domains, whereas socioemotional awareness showed less pronounced differences. This pattern is consistent with previous studies suggesting that awareness in AD is multidimensional, with cognitive and functional deficits recognized less frequently than socioemotional changes.^37–39

The absence of significant differences between groups in the domain related to awareness of social functioning suggests that this domain of awareness may be more resistant to decline or influenced by more stable psychosocial and emotional coping mechanisms.^8,18,19 In the ASPIDD, socioemotional awareness refers to the individual's ability to perceive and appropriately respond to social and emotional cues, such as recognizing others’ feelings and maintaining interpersonal sensitivity. This domain is relevant to decision-making capacity because effective choices often require not only cognitive reasoning, but also an awareness of the social and emotional context in which decisions are made. Thus, awareness is not a unitary construct that declines linearly, rather, it is comprised of different domains, each with its own pattern of decline. The pattern of impairment in the domains of awareness may be influenced by different clinical aspects related to the illness.¹⁸ Preservation of social and emotional awareness even in more advanced stages of AD highlights the importance of including these aspects in the treatment and care of patients.

Regression analyses revealed that the predictors of decision-making capacity varied with the advance of the disease. In people with MCI, all domains were significantly influenced by the presence of neuropsychiatric symptoms, indicating that behavioral and emotional symptoms can impair the domains of decision-making capacity, even in early stages of the disease. Neuropsychiatric symptoms may affect and interfere with cognitive processes that are crucial for decision-making capacity.^13,40,41 Individuals with MCI, whose ability to express a choice was influenced by cognitive factors, support the assumption that this area is more susceptible to cognitive load, as it relies on cognitive resources for understanding, appreciating, and making judgments. In individuals with AD, neuropsychiatric symptoms remained a significant predictor of most domains, and functionality was also associated with expression of choice. In moderate AD, decision-making capacity domains were influenced by a wider range of factors: neuropsychiatric symptoms, awareness, especially in the understanding and appreciation domains, and functionality, particularly in the expression of choice domain. These findings support previous studies, showing that decision-making capacity is a multifactorial construct influenced not only by cognition but also by behavioral, emotional, and functional state.^8,42,43

Emotions modulate cognitive processes and affect perceptions of costs and benefits, influencing choices. Likewise, behavioral factors can impair decision-making capacity. Therefore, a comprehensive assessment of decision-making capacity must consider this interaction between cognitive, emotional, and functional elements, not just cognitive functions.^41,44,45 In individuals without neuropsychiatric symptoms, declines in awareness and decision-making capacity tend to progress in a cognitively driven and more linear manner, with relative preservation of appreciation and expression of choice in early stages. In contrast, neuropsychiatric symptoms introduce additional variability and disproportionately compromise decision-making capacity, underscoring the need for more targeted investigation of their joint trajectory.

This study has some limitations. The cross-sectional design, which did not allow for an analysis of the progression of awareness or decision-making capacity. A longitudinal study would allow for observing the variations and the relationship between awareness and decision-making capacity over time. Diagnosis of individuals with AD and MCI was established based on clinical examination, medical history, neuroimaging, and neuropsychological assessments; biomarkers were not used in the diagnostic process. In addition, the sample was recruited at an outpatient center for dementia, which hinders generalization of the results to people with MCI and AD who are not receiving such specialized clinical support. Another limitation of this study was the limited scope of cognitive testing for executive impairment, which has been implicated in instrumental activities of daily living (IADL) decline. The absence of more detailed cognitive measurements, and the use of the MMSE as the sole cognitive test, restrict our ability to fully explore this mechanism. Future studies should examine whether these associations extend to decision-making processes, which could further elucidate the role of executive dysfunction in everyday functioning. This study serves as a foundation for future research. Further studies with larger sample sizes are needed to gain a deeper understanding of the role of awareness in decision-making capacity in individuals with MCI and AD. Future studies could use biomarkers such as amyloid PET or CSF to establish the diagnosis, due to recent updates to NIA-AA criteria for diagnosis and staging of AD.

Clinical implications

Clinically, our findings suggest that decision-making capacity in individuals with MCI and AD should be assessed regularly to ensure that it remains appropriate for each specific situation.

Healthcare professionals must consider an individual's cognitive capacity and level of awareness to ensure informed and appropriate medical decisions.^42,46 Understanding the specific challenges faced by people with MCI and AD in decision-making can help healthcare providers adapt treatment plans aligned with the individuaĺs abilities and preferences. It may involve more collaboration with family and caregivers to ensure that the patient's needs and wishes are considered. This knowledge can influence how support systems are designed in case of patient-care agreements. It might lead to more comprehensive and flexible care plans that adapt to changes in the individuals’ cognitive function. Normally, caregivers underestimate the decision-making capacity of the person with AD and MCI,⁴⁷ as they are impacted by the burden, depression, the patient's functional ability, and the type of relationship.⁴⁸ Currently, there are discussions about using tools, such as advanced directives, which determine the wishes and preferences of the person with MCI and AD based on prior knowledge of values and preferences that can help in decision-making capacity.^49–53

Additionally, an important direction for future research should be mentioned. In particular, it would be valuable to investigate whether patients’ decision-making capacity improves once underlying neuropsychiatric symptoms are more effectively managed, which could help clarify the extent to which these symptoms exert a potentially reversible influence on decisional autonomy. It would also be important to examine whether these relationships differ across other dementia etiologies, such as frontotemporal dementia, Lewy body dementia, and vascular dementia, given that distinct clinical and neuropsychiatric profiles may affect decision-making capacity in different ways.

Conclusion

Individuals with MCI showed greater preservation of awareness of their cognitive functioning and health status than those with mild and moderate AD. In addition, awareness played an essential role in the understanding domain of decision-making capacity. Clinically, our findings shed light on the need to consider differences in the domains of decision-making capacity and their association with clinical aspects such as awareness according to the diagnosis. Furthermore, our data can suggest hypotheses for larger, more robust, prospective studies.

Footnotes

Acknowledgements

The authors acknowledge the financial support provided by Brazilian funding agencies: Natalie Aparecida Pereira de Souza received support from the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES); Marcela Moreira Lima Nogueira and Tatiana Belfort Almeida dos Santos received support from the Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ); and Márcia Cristina Nascimento Dourado is a Researcher of the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and also received support from FAPERJ.

ORCID iDs

Natalie Aparecida Pereira De Souza

Marcia Cristina Nascimento Dourado

Ethical considerations

This study was approved by the Ethics Committee of the Institute of Psychiatry at the Federal University of Rio de Janeiro and was conducted in accordance with the Declaration of Helsinki.

Consent to participate

Written informed consent was obtained from all caregivers and from the people with AD or, when applicable, from their legal guardians/family representatives.

Consent for publication

Not applicable.

Author contribution(s)

Natalie Aparecida Pereira De Souza: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Writing – original draft; Writing – review & editing.

Tatiana Teresa Belfort Almeida Dos Santos: Methodology; Writing – review & editing.

Marcela Moreira Lima Nogueira: Investigation; Writing – review & editing.

Marcia Cristina Nascimento Dourado: Conceptualization; Data curation; Funding acquisition; Methodology; Project administration; Resources; Supervision; Validation; Visualization; Writing – review & editing.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study received support from CAPES, FAPERJ, and CNPq, as detailed in the Acknowledgments section.

Declaration of conflicting interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Marcia Cristina Nascimento Dourado is an Editorial Board Member of this journal but was not involved in the peer-review process of this article nor had access to any information regarding its peer-review.

Data availability statement

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

References

Karlawish

Grill

. Alzheimeŕs disease biomarkers and the tyranny of treatment. EBioMedicine 2024; 108: 105291.

Darby

Dickerson

. Dementia, decision making, and capacity. Harv Rev Psychiatry 2016; 25: 270–278.

Appelbaum

Grisso

. Assessment patientś capacities to consent to treatment. N Engl J Med 1988; 319: 1635–1638.

Grisso

Appelbaum

. MacArthur Competence Assessment Tool for Treatment (MacCAT-T). Sarasota, FL: Professional Resources Press, 1998.

Brand

Schiebener

. Interactions of age and cognitive functions in predicting decision-making under risky conditions over the life span. J Clin Exp Neuropsychol 2016; 35: 9–23.

Geng

Wang

, et al. Reduced delayed reward selection by Alzheimeŕs disease and mild cognitive impairment patients during intertemporal decision-making. J Clin Exp Neuropsychol 2020; 42: 298–306.

Parmigiani

Del Casale

Mandarelli

, et al. Decisional capacity to consent to treatment and research in patients affected by mild cognitive impairment. A systematic review and meta-analysis. Int Psychogeriatr 2022; 34: 529–542.

De Souza

NAP

de Oliveira

de Carvalho

RLS

, et al. The relationship between decision-making capacity and the domains of awareness in Alzheimer disease. Alzheimer Dis Assoc Disord 2022; 36: 58–63.

Zhang

Chen

Dong

, et al. Autonomous decision-making of UAV cluster with communication constraints based on reinforcement learning. J Cloud Comp 2025; 14: 12.

10.

Menne

Tucke

Whitlatch

, et al. Decision-making involvement scale for individuals with dementia and family caregivers. Am J Alzheimers Dis Other Demen 2008; 23: 23–29.

11.

Clare

. The construction of awareness in early-stage Alzheimeŕs disease: a review of concepts and models. Br J Clin Psychol 2004; 43: 155–175.

12.

Dourado

Mograbi

Santos

, et al. Awareness of disease in dementia: factor structure of the assessment scale of psychosocial impact of the diagnosis of dementia. J Alzheimers Dis 2014; 41: 947–956.

13.

Lacerda

Baptista

MAT

Belfort

, et al. Measures of awareness in Alzheimer’s disease and clinical differences between groups with preserved, impaired, and absent awareness. J Alzheimers Dis Rep 2024; 8: 399–410.

14.

Piras

Orfei

Caltagirone

, et al. Self-awareness in mild cognitive impairment: quantitative evidence from systematic review and meta-analysis. Neurosci Biobehav Rev 2016; 61: 91–107.

15.

Stites

Dupont

Herades

, et al. Conscience des troubles dans la maladie d´Alzheimer et le mild cognitive impairment [awareness disorders in Alzheimeŕs disease and in mild cognitive impairment]. Int J Geriatr Psychiatry 2014; 40: 180–187.

16.

Jacus

. Awareness, apathy, and depression in Alzheimeŕs disease and mild cognitive impairment. Brain Behav 2017; 7: e00661.

17.

Petersen

RCF

Lopez

Armstrong

, et al. Practice guideline update summary: mild cognitive impairment [RETIRED]: report of the guideline development, dissemination, and implementation subcommittee of the American academy of neurology. Neurology 2018; 90: 126–135.

18.

Baptista

MAT

Lacerda

Belfort

, et al. Neuropsychiatric symptoms and awareness in young-and-late-onset dementia. Aging Ment Health 2025; 29: 1981–1988.

19.

De Souza

NAP

Simões

JPN

Dourado

MCN

. The relationship between decision-making capacity and awareness in people with young-onset Alzheimer’s disease. Alzheimers Dis 2024; 101: 671–680.

20.

Faul

Erdfelder

Lang

, et al. G*power 3: flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 2007; 39: 175–191.

21.

American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V). Washington, DC: American Psychiatric Association, 2013.

22.

Morris

. The clinical dementia rating (CDR): current version and scoring rules. Neurology 1993; 43: 2412–2414.

23.

Maia

Godinho

Ferreira

, et al. Application of the Brazilian version of the CDR scale in samples of dementia patients. Arq Neuropsiquiatr 2006; 64: 485–489.

24.

Santos

Sousa

MFB

Simões

JPN

, et al. Macarthur competence assessment tool for treatment in Alzheimer disease: cross-cultural adaptation. Arq Neuropsiquiatr 2017; 75: 44–50.

25.

Grisso

Appelbaum

Hill-Fotouhi

. The MacCAT-T: a clinical tool to assess patient´s capacities to make treatment decisions. Psychiatr Serv 1997; 48: 1415–1419.

26.

Folstein

McHugh

. Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12: 189–198.

27.

Bertolucci

PHF

Brucki

SMD

Campacci

, et al. Mini-Exame do estado mental em uma população geral: impacto da escolaridade. Arq Neuropsiquiatr 1994; 52: 01–07.

28.

Pfeffer

Kurosaki

Harrah

, et al. Measurement of functional activities in older adults in the community. J Gerontol 1982; 37: 323–329.

29.

Dutra

MCC

Lima

Bertolucci

PHF

, et al. Accuracy and reliability of the pfeffer questionnaire for the Brazilian elderly population. Dement Neuropsychol 2015; 9: 28–35.

30.

Cummings

Mega

Gray

, et al. The neuropsychiatric inventory: comprehensive assessment of psychopathology in dementia. Neurology 1994; 44: 2308–2314.

31.

Camazzotto

Kochhann

Simeoni

, et al. Reliability of the Brazilian Portuguese version of the neuropsychiatric inventory (NPI) for patients with Alzheimeŕs disease and their caregivers. Int Psychogeriatr 2008; 2: 383–393.

32.

Alexopoulos

Abrams

Young

, et al. Cornell scale for depression in dementia. Biol Psychiatry 1988; 23: 271–284.

33.

Portugal

Coutinho

ESF

Almeida

, et al. Validation of montgomery-asberg rating scale for depression in dementia in Brazilian elderly patients. Int Psyshogeriatr 2012; 224: 1291–1298.

34.

Grisso

Appelbaum

. Comparison of standards for assessing patientś capacities to make treatment decisions. Am J Psychiatry 1995; 152: 1033–1037.

35.

Marson

Hawkins

McInturff

, et al. Cognitive models that predict physician judgments of capacity to consent in mild Alzheimer’s disease. J Am Geriatr Soc 1997; 45: 458–464.

36.

Sun

Xie

Wang

, et al. Decision-making under ambiguity or risk in individuals with Alzheimer’s disease and mild cognitive impairment. Front Psychiatry 2020; 18: 11–28.

37.

Belfort

Brandt

Silva

, et al. Comparing patterns of impairment in social cognition between young-onset and late-onset Alzheimer’s disease. J Alzheimers Dis 2025; 75: 1057–1066.

38.

Ayodele

Rogaeva

Kurup

, et al.

Early-onset Alzheimer’s disease: what is missing in research?

Curr Neurol Neurosci Rep 2021; 21: 4.

39.

Lacerda

Baptista

MAT

Belfort

, et al. Perception of change in sexual activity in Alzheimeŕs disease: views of people with dementia and their spouse-caregivers. J Geriatr Psychiatry Neurol 2016; 29: 267–273.

40.

Koçak

FOK

Kumral

. The relationship between dementia staging scales, cognitive-behavioral scales and functionality in patients with cognitive impairment. PLoS One 2025; 20: e0322572.

41.

Alfeo

Lanciano

Abbtantuono

, et al. Cognitive, emotional, and daily functioning domains involved in decision-making among patients with mild cognitive impairment: a systematic review. Brain Sci 2024; 14: 278.

42.

Palmer

Harmell

. Assessment of healthcare decision-making capacity. Arch Clin Neuropsychol 2016; 31: 530–540.

43.

Moye

Marson

. Assessment of decision-making capacity in older adults: an emerging area of practice and research. J Gerontol B Psychol Sci Soc Sci 2007; 62: P3–P11.

44.

Brosch

Scherer

Grandjean

, et al. The impact of emotion on perception attention, memory, and decision-making capacity. Swiss Med Wkly 2013; 143: w13786.

45.

Sarmiento

EAR

Sarmiento

GNR

Sr Sarmiento

NVR

Sr , et al. Usability assessment in an E-health app for cognitive stimulation (LABPSI) in elderly population. Alzheimers Dement 2025; 20: e092355.

46.

Iriniparast

Shi

, et al. Cognitive reserve and mild cognitive impairment: predictors and rates of reversion to intact cognition vs progression to dementia. Neurology 2022; 98: e1114–e1123.

47.

van Duinkerken

Farme

Landeira-Fernandez

, et al. Medical and research consent decision-making capacity in patients with Alzheimer’s disease: a systematic review. J Alzheimers Dis 2018; 65: 917–930.

48.

Mattos

Goodarzi

Veroniki

, et al. Comparative efficacy of interventions for aggressive and agitated behaviors in dementia. Ann Intern Med 2023; 171: 663–642.

49.

Watt

Goodarzi

Veroniki

, et al. Comparative efficacy of interventions for aggressive and agitated behaviors in dementia. Ann Intern Med 2019; 171: 633–642.

50.

Porteri

Ienco

Turla

, et al. Shared care planning in people with cognitive disorders and dementia: a survey among patients and caregivers in Italy. BMC Med Ethics 2024; 25: 145.

51.

Stern

. Cognitive reserve in ageing and Alzheimer’s disease. Lancet Neurol 2012; 11: 1006–1012.

52.

Toschi

Devanarayan

, et al. The age-specific comorbidity burden of mild cognitive impairment: a US claims database study. Alzheimers Res Ther 2023; 15: 211.

53.

Ding

. Compromised agency in the subjective experiences of temporal distortions among institutionalized Alzheimer’s patients in China: a qualitative study. Qual Health Res 2025; 36: 725–738.