Relationship between parental history of dementia, motor-cognitive and executive function performance in African American women

Abstract

People with parental history (PH) of Alzheimer's Disease (AD) and Alzheimer's Disease and related dementias (ADRD) are themselves at risk of developing dementia. ADRD are more prevalent in African Americans and women. A decline in executive function and motor-cognitive integration can cause an impaired performance of functional skills. The monitoring of cognitive and psychosocial function in individuals with a PH of ADRD is important for implementing interventions to delay or prevent ADRD diagnosis. This study compared 58 African American women (M age = 63.2 ± 7.2 years) with PH of ADRD (n = 34) versus without PH (NPH; n = 24) on the performance of motor-cognitive and executive function tasks, and mental and physical quality of life (QOL) using point biserial correlations and linear regression. Linear regression revealed no difference between participants with and without PH on motor-cognitive tests. However, compared to participants with NPH, participants with PH of ADRD performed significantly worse on the DKEFS (Delis Kaplan Executive Function System) Tower Test (PH: M = 9.9 ± 2.0; NPH: M = 11.5 ± 4.3; p = 0.046), had poorer mental QOL (PH: M = 46.8 ± 10.7; NPH: M = 52.8 ± 7.8 l; p = 0.007); and physical QOL (PH: M = 40.9 ± 9.3; NPH: M = 44.7 ± 8.6; p = 0.023). African American women at risk for ADRD may exhibit deficiencies in executive function and physical and mental quality of life before memory deficits meet the criterion for ADRD diagnosis. Motor-Cognitive tasks may be preserved. Executive function and mental and physical health-related QOL may be important targets for identifying individuals at increased risk for ADRD and developing appropriate rehabilitative interventions.

1. Introduction

Dementia is defined as cognitive decline that severely affects daily functioning. The most common cause of dementia is Alzheimer's Disease (AD). Approximately 50 million people are living with Alzheimer's Disease and Related Dementias (ADRD) with almost 10 million new cases every year [1]. In the early stages of AD, several years of decline precede medical attention [2]. Increasing prevalence and a prolonged period of decline, particularly among African Americans [3], makes it imperative to contribute to the body of research, developing better diagnostic measures and more efficient screening procedures for those in early disease stages [4,5]. With early ADRD diagnosis, individuals can participate in their own long-term planning and take advantage of both pharmacological and non-pharmacological treatments that can help delay cognitive deterioration [6].

The interaction of many environmental, sociocultural, behavioral, and biological factors over the life course are implicated in the development of ADRD [7]. Biologically, having a first-degree relative with ADRD increases an individual's risk for developing ADRD by six times [8]. The apolipoprotein E gene (APOE) ε4 allele, is a recognized genetic risk factor for AD in both men and women [9]. Having the APOE ε4 allele and having a family history is associated with developing AD [8,10,11]. However, independent of genetic factors including the APOE ε4 allele, parental history (PH) is still associated with dementia risk [12]. Increased ADRD risk among those with PH occurs within a historical epoch of racism and other “isms,” and may be influenced by shared familial environmental exposures, sociocultural expectations and opportunities, and behavioral and lifestyle factors contributing to ADRD risk (i.e., sharing a living environment and eating a similar diet as other members in the household) [13]. Biological epigenetic changes in gene expression based on the individual's exposome may also increase ADRD risk, although the data collected on this subject are insufficient to suggest discrete epigenetic pathways to ADRD [14,15].

Race and ADRD have a complex, non-causal relationship impacted greatly by experiences with systemic racism and race-based biases against non-White individuals. Statistically, African Americans with a family history of ADRD by a first-degree biological relative have a higher cumulative risk of development of dementia than White Americans with family history [8]. ADRD are 64% more prevalent in African Americans than in White Americans [3]. Despite these statistics, there is insufficient data to establish causal pathway links between racism, race, and ADRD, as the data on biological drivers of ADRD among African Americans and studies on how biology and lived experience culminate in disease are sparse [16]. While neuro-biological factors may contribute to some observed differences in ADRD prevalence by race, the higher AD and ADRD prevalence among African Americans could be largely attributable to individual and structural racism, discrimination, transgenerational trauma [14,15], and environmental factors [17].

Race is a socially constructed classification system usually denoting differing shades in skin color and national origin [16,18]. Despite this arbitrary categorization, the health impacts of being classified as “non-white” in American society are real. The true relationship between African Americans and increased ADRD risk is possibly that the negative impacts of disparate socioeconomic opportunity and socio-cultural disadvantages caused by institutional racism may cause African Americans to have increased stress and depressive symptoms. The resultant higher allostatic load (cumulative impact of stressors and negative exposures and experiences) [19] may contribute to the development of ADRD and may lead to transgenerational increased risk for ADRD among African Americans [20–22].

Racism-related racial disparities in ADRD extend beyond risk and prevalence. Neuropathology from an autopsy can be more evident in African Americans than White Americans [23]. Multiple studies indicate that neurofibrillary tangle load, a potential cause for ADRD, is greater in African Americans compared to White Americans [24–26]. Unfortunately, African Americans with ADRD typically exhibit greater symptom severity and are less likely to receive treatment than White Americans [27,28]. As with the increased risk and prevalence for ADRD experienced by African Americans, the more severe ADRD symptomology among African Americans suggests a complex interplay of the impacts of racism and biological-environmental interactions [27,28] compounded by systemic disparities in access to dementia care [29], socioeconomic disadvantages [30], and higher allostatic load often observed among African Americans [19].

Biological sex is another complicated factor in the development of ADRD. Women are nearly twice as likely to develop ADRD as men [31]. Women have some elevated risk for ADRD because of their longevity, since greater age is associated with cognitive decline, and women tend to live longer than men [32,33]. Female hormone-related changes, including earlier age of onset of menopause [34,35] hysterectomy [36], pregnancy, and hormone therapy have been shown to influence cognitive function and to play roles in the development of ADRD [37,38].

African American women have the highest estimated allostatic loads compared to African American men and White individuals, [39,40]. Allostasis can have a tremendous impact on the development of ADRD [20–22,41]. The higher allostatic load among African American women could be attributed to sex-specific experiences. African American women often experience racial discrimination which can contribute to internal “superwoman schemas” also called the “strong Black woman” schema [42], where African American women feel pressure to over-compensate for negative stereotypes [43]. Further, sociocultural gender role expectations dictate that African American women care for others, including children and older adults, as these responsibilities have been relegated to African American women since the era of race-based slavery when caring was assigned to African American women [44], to the present day, given high incarceration rates of African American men and their subsequent unavailability to assist in caregiving responsibilities from behind bars [45]. Caregiving is often quite stressful [46]. This may explain why women carriers for the APOE ε4 allele have a greater risk than men of transitioning from normal cognition to mild cognitive impairment (MCI), and from MCI to ADRD (i.e., the basis for this transition may be related to allostatic load rather than gene inheritance) [47] and why neurofibrillary tangles may also develop more in women than in men [48].

1.1. Disease trajectories

Experience with cognitive decline varies between individuals regardless of race, racism, or sex [49–51]. Cognitive decline can occur secondary to different disease pathways and can impact different brain areas, resulting in a variety of cognitive, affective, and behavioral challenges [52,53]. In amnestic MCI, memory is the primary domain impacted, compared to non-amnestic MCI, in which thinking is the primary domain impacted [54]. Amnestic MCI is generally a precursor to ADRD, yet as the disease progresses, non-memory domains of cognition also become impaired, resulting in multimodal MCI in which both memory and thinking ability decline [55]. “Dementia” refers to the point in disease progression where both remembering and thinking abilities have declined to the point where the individual is no longer able to complete activities of daily living independently [56,57].

1.2. Executive function

One of the domains that can be impacted early on during the ADRD trajectory is executive function. Executive function includes abilities such as planning, problem-solving, cognitive flexibility, and performing goal-oriented activities [58]. There is a growing body of evidence that amnestic and non-amnestic MCI can be preclinical signs for ADRD such that impaired executive function among those with MCI can be an early indicator for future progression to ADRD [59]. Executive function decline was shown to precede average memory decline in aging adults [59–61]. Beta-amyloid accumulation was associated with worse executive function in adults without memory problems [62]. Early executive dysfunction might reflect subtle, early changes in persons at risk of late-onset AD [63]. In a longitudinal study conducted over 1.5 years, people diagnosed with AD experienced decreased executive functioning [64]. A study that included mostly male participants with a family history of AD performed worse on executive function tests than those without family history [65]. A small study with 18 older adults with a family history of AD performed worse on the Wisconsin Card Sorting Test, a measure of executive function via assessing response inhibition, abstract reasoning, and perseveration in comparison with age and education matched participants without a family history [66].

1.3. Motor cognitive function and visuospatial function

Motor-cognitive integration refers to activities requiring concurrent action and thought. These dual tasks comprise broad circuitry including the intraparietal sulcus, left dorsal premotor cortex, right anterior insula, and frontoparietal clusters, and research continues to investigate neurological underpinnings [67]. Due to AD progression, progressive decline in Instrumental Activities of Daily Living (IADLs), e.g., meal preparation and household chores [68], are often noted in individuals in the early stages of AD [2]. Persons with MCI and AD have more difficulty performing dual tasks, such as these IADL’s like household chores, meal preparation, driving, that require both motor and cognitive skills, than healthy older adults [69]. Paramount in motor-cognitive tasks are deficits in divided control. As cognition worsens motor-cognitive deficits become more pronounced and have been shown to be associated with progression to dementia [70,71]. Persons with MCI and AD have more difficulty performing dual tasks, such as these IADL’s like household chores, meal preparation, driving, that require both motor and cognitive skills, than healthy older adults [69]. Paramount in motor-cognitive tasks are deficits in divided control. As cognition worsens motor-cognitive deficits become more pronounced and have been shown to be associated with progression to dementia [70,71]. A small study examining primarily White participants indicated that, compared to non APOE ε4 carriers, APOE ε4 carriers were more impaired at performing a task requiring both motor and cognitive skill even though the two groups were not different in global cognition [72]. Further, visuo-spatial function and motor-cognitive integration are crucial for navigation and mobility required to perform IADLs [73,74].

1.4. QOL

Individuals with ADRD experience a range of neuropsychiatric symptoms which can contribute to decreased QOL [75,76]. Additionally, functional challenge in individuals with dementia is positively associated with reduced QOL [77]. This can contribute to losing the ability to perform functional skills [70,71,78], impacting QOL. Therefore, reduced QOL may also be a precursor to greater loss of function throughout the ADRD disease course. Little has been studied about QOL in individuals with PH of dementia. Yet, caregivers caring for a family member often have reduced QOL compared to a normative, age-adjusted sample [79,80].

Few studies investigate ADRD in African Americans and women despite their increased risk and potential differences with Whites and men secondary to inequitably distributed social determinants of health [27,81]. In this pilot study we aimed to investigate executive function, motor cognitive integration, and QOL in 58 African American women with parental history (PH) and without parental history (NPH) of ADRD.

We hypothesized that those with PH would exhibit greater deficits across motor-cognitive, executive function, and QOL measures compared to NPH. The data gathered from this inquiry may be used to power a future longitudinal study that will identify a time of initial decline when high-risk patients (based on sex, increased cognitive decline risks secondary to racism leading to race-based inequities in socio-economic determinants of health, and genetics) could begin implementing non-pharmacological interventions addressing inequities in housing, education, and other social determinants of health aiming to reduce allostatic load, improve support of the diagnosed, and promote protective lifestyle decision-making such as diet, exercise, and coping strategies in order to protect against the continued progression of ADRD. While hypothesis testing was performed in this study, the findings presented here are intended to primarily serve as effect sizes for a future more definitive and adequately powered trial in this important and underserved demographic.

2. Materials and methods

Informed consent was obtained from all participants prior to study entry. The Institutional Review Board of Emory University and the Atlanta VA R&D Review committee approved this protocol.

2.1. Participants

Participants were recruited from the Emory Alzheimer's Disease Research Center and Wesley Woods Health Center. These locations are sites for frequent research collaborations. There is an overlap in the patient populations because both sites address the needs of geriatric people with neurodegenerative disease. Cross-sectional, secondary retrospective data analyses were performed on baseline data collected from African American women who were at least 45 years of age and enrolled in longitudinal rehabilitation and behavioral studies conducted from 2014 to 2019. The parent study protocol is outlined in detail elsewhere [Blinded]. All participants were middle to upper class, lived in the Atlanta metro area and surrounding communities and identified as “Black/African American.” English was the primary language of all participants. Women with pre-existing diagnosed ongoing mental/physical conditions including cancer and severe depression were excluded. Fifty-eight African American women with (PH: n = 34) and without (NPH: n = 24) PH of ADRD were included. ADRD diagnosis of the parent was verified using the validated Dementia Questionnaire (DQ) which includes questions about memory, behavior changes, cognitive function, and difficulties performing IADLS and thus screens for non-type specific dementia [82] and medical records when available. Participants' caregiver status refers to if a participant was the primary caregiver of their parent with ADRD (yes or no). Data on other types of caregiving activities were not collected.

2.2. Measures

Clinical characteristics and demographics of participants were collected including age, primary language, education, occupational status, body mass index (BMI), hypertension status, number of falls in the past year, fall worry, marital status, assistive device use, and number of times leaving the house in a week, and caregiver status when available. Cognitive and motor testing took approximately 120 min and was completed at a university health center.

Research indicates that motor measures are more predominantly associated with white matter hyperintensities, a part of the pathophysiology of motor decline, in persons with ADRD than with individuals with mild cognitive impairment [83]. Additionally, research has shown that individuals with risk factors for AD development performed worse on dual tasks than memory tasks [72]. Because impairments in global cognitive and motor function independently have not always been detected at earlier, preclinical stages of ADRD, global cognition (Montreal Cognitive Assessment) and motor measures (Timed Up and Go) were examined for their association with PH.

The following measures were used in the battery because they are validated clinical measures commonly used in populations at risk for ADRD.

2.2.1. Global cognition measure

Montreal Cognitive Assessment (MoCA) is a screening tool for MCI [84]. It includes eight cognitive domains assessing clock drawing and 3-D cube copy task, Trail Making B task, phonemic fluency task, two-item verbal abstraction task, sustained attentional task, serial subtraction task, naming task, repetition of sentence task, and orientation to time and place task.

2.2.2. Motor measure

Timed Up and Go (TUG) was used to examine mobility in patients by timing participants as they stand up, walk 3 m away, turn around and go back to sit in the chair [69].

2.2.3. Motor cognitive integration

Since motor-cognitive integration is key to performing activities of daily living, multiple measures of motor-cognitive function and integration were administered to increase ecological validity of findings.

Timed Up and Go Cognitive (TUG-cognitive) was administered in addition to the TUG to investigate motor-cognition in people susceptible to Alzheimer's disease [69]. While completing the TUG protocol (as described above), participants were asked to count backwards by 3's (serial 3's) from a random number between 20 and 100. The TUG-Cog dual task cost (Percent) is equal to the time (sec) for TUG Cognitive minus TUG (sec), divided by TUG (sec) and multiplied by 100. Dual task challenges including TUG-cognitive tests have been shown to have clinical utility for detecting early-stage dementia by predicting likelihood of MCI-ADRD transition (OR) = 4.06, 95% confidence interval (CI) 2.28–7.23, p < 0.001) [85,86].

Four Square Step Test (FSST) is used to test standing balance and mobility by timing participants as they step clockwise and then counterclockwise in four squares formed by shower curtain rods [87]. FSST performance is indicative of motor-cognitive integration, which is impaired in many individuals with ADRD [87–89]. Participants were asked to perform the task without touching the rods and timing began as soon as the participant-initiated movement. The time to complete the movement was recorded. Body Position Spatial Task (BPST) tests whole body visuospatial functioning and incorporates steps and turns in a specified order through modeling the researcher [90]. In this way, the integration of cognition and motor function was investigated. The product of the span (highest number of correctly repeated steps) and number of correct trials completed was calculated.

2.2.4. Executive function measures

Executive function was evaluated using standard clinical measures, including the D-KEFS (Delis Kaplan Executive Function System) Color Word Interference Task (CWIT), and the D-KEFS Tower task, in which individual performance on tasks was converted into scaled values based on age group norm, per the validated D-KEFS scoring manual [91,92]. These tests were selected to measure executive function since D-KEFS is a nationally standardized assessment battery of executive function [91,92].

Through four conditions, the Color-Word Interference Test measures inhibition and inhibition and switching, aspects of executive function [93]. The first condition, the Color Naming task, involves a participant saying the names of the colors of a series of red, green, and blue squares. In the second condition, the Word Reading task, the participant is asked to read the words “red,” “green,” and “blue” printed in black ink. The inhibition trial requires a participant to say the name of the color of ink of words “red,” “green,” and “blue” that is printed in incongruent ink colors. The inhibition/switching trials involve a page with “red,” “green,” and “blue” printed in either red, green, or blue ink and some words are printed inside boxes. Participants must say the color of the ink of words inside boxes. Time for each trial, total number of errors, both uncorrected and self-corrected, were recorded converted into age-normed subscales and scores per published methods [92]. For all variables but contrast scores, higher scores indicate better performance.

D-KEFS Tower Test (DTT) was used to assess the planning and organization aspects of executive function [94]. Participants move three rings of different sizes on pegs in specific arrangements printed on cards that were presented by the administrator. They were told to attempt to make the arrangement in the least amount of moves possible. The number of moves as well as time taken were recorded and converted into subscales and scores per published methods. [92]

Trails A and B is a test of visuo-perceptual, working memory, and executive functioning [95]. In part A, participants were asked to connect numbers 1–25 in sequential order as fast as possible. In part B, participants alternate between connecting numbers (1–13) and letters (A-L) in order. The time to complete both trials was recorded and the difference score between A and B was calculated by subtracting the time to complete A from the time to complete B.

2.2.5. Questionnaires

The Short Form Health Survey-12 (SF-12) is comprised of twelve questions regarding how the individuals feel and how well they can complete their usual activities. The scores for the subscales, Physical and Mental Composite Scales were considered in analyses. All subscales were calculated according to published methods [96].

2.3. Statistical analysis

Descriptive statics were calculated and compared between PH and NPH groups using Fisher's exact tests and independent t-tests as appropriate. Point-biserial correlations were used to examine associations between the PH status (PH vs. NPH) and participant performance on global cognitive, motor, motor-cognitive, and executive function measures. Cohen's conventions were used to determine the strength of association where 0.1 is small, 0.3 is moderate, and 0.5 is large. Linear regression analysis with a p-value calculation was used to predict performance on the various measures based primarily on PH or NPH status. Two models were used. The number of times leaving the house weekly (life span) differed between the groups and was therefore used as a covariate in the first model. The second regression model adjusted for number of times leaving the house weekly and included two variables of health of import in AD and African American women: body mass index (BMI), and hypertension diagnosis to account for the effects of other potentially influencing variables [97]. The α level was set at 0.05. All analyses were carried out using R software (version 1.2.1335).

3. Results

Fifty-eight African American women were included in this study (age = 63.2 ± 7.2 years, range: 48–81 years). 34 participants had a PH of ADRD, and 24 participants did not. Of those with PH, 16 were the primary caregiver of their parent with AD. Collecting data on caregiver status was added after 17 individuals had already completed assessments, and as such, caregiving status is not available for these individuals. “Sandwich,” (multi-generational care provision) child, and other types of informal caregiving status were not collected on any participant. Descriptive and demographic statistics of the sample are summarized in Table 1. Participants did not differ significantly in age, education level, occupational status, number of falls in the past year, fear of falling, and marital status. Participants with PH showed a trend of leaving the house fewer times per week.

Table 1.

Characteristics of the sample including individuals with and without parental history of ADRD.

	Entire sample (n = 58) Mean (SD)/N (%)	Parental history (n = 34) Mean (SD)/N (%)	No parental history (n = 24) Mean (SD)/N (%)	P-Values
Agea (years)	63.2 (7.3)	61.7 (7.4)	65.2 (6.7)	0.069
Year of Educationa	14.2 (2.4)	13.8 (2.5)	14.5 (2.3)	0.265
Montreal Cognitive Assessment (/30)a	25.2 (3.1)	24.7 (3.5)	25.6 (2.8)	0.286
PASE Scorea	124.5 (60.8)	113.5 (59.2)	132.7 (61.7)	0.245
Occupational Statusb*
Work full-time	15(25.9)	10(29.4)	5(18.5)
Work part-time	8(13.8)	4(11.8)	4(14.8)
Homemaker	1(1.7)	0(0)	1(3.7)
Retired	29(50.0)	16(47.1)	13(55.6)
Unemployed/seeking work	1(1.7)	1(2.9)	0(0)
Disabled	4(6.9)	3(8.8)	1(7.4)	0.768
Body Mass Indexa (kg/m2)	30.0 (5.6)	30.0 (5.1)	30.2 (6.4)	0.844
Hypertension
Yes	39(67.2)	22(64.7)	17(70.8)
No	19(32.8)	12(35.3)	7(29.2)	0.778
Number of Falls in the Past Yeara	0.6(1.7)	0.5(0.9)	0.7(2.4)	0.734
Fall Worryb
Not at All	29(50.9)	17(51.5)	12(50.0)
A Little	22(38.6)	12(36.4)	10(41.7)
Moderately	6(10.5)	4(12.1)	2(8.3)
High	0(0)	0(0)	0(0)	0.225
Marital Statusb
Single	5(8.8)	2(5.9)	3(13.0)
Married/Partnered	22(38.6)	12(35.3)	10(43.5)
Divorced	19(33.3)	15(44.1)	4(17.4)
Widowed	11(19.3)	5(14.7)	6(26.1)	0.141
Assistive Device Useb
Yes	4(6.9)	2(5.9)	2(8.3)
No	54(93.1)	32(94.1)	22(91.7)	p > 0.99
Number of Times Leaving House per Weekb
4 Times per Week or Fewer	25(43.1)	11(41.7)	14(58.3)
Everyday	33(56.9)	23(67.6)	10(41.7)	0.063

^aTwo-tailed, independent T-Tests were used for continuous variables.

^bFisher’s exact tests were used for categorical variables.

^*P values indicate significant differences between Parental History and No Parental History Groups at the 0.05 level.

3.1. Point biserial correlations

Using a 95% Confidence Interval and per Cohen's conventions, worse scores on the SF-12 MCS (mental composite score) were significantly and moderately correlated (r = −0.302) [−0.520 – −0.048] with having PH. Worse scores on the SF-12 PCS (physical composite score) were moderately but not significantly correlated (r = 0.200) with PH. Worse scores on the DTT Total Achievement Scaled Score (age-normed according to nationally-validated D-KEFS scoring conventions) were moderately correlated (r = 0.256) [−0.483–0.002] with PH although this correlation was not significant. Small or weak associations (r = 0.1–0.2) were observed for all other variables (Table 2).

Table 2.

Point biserial correlations between parental history and performance on motor, cognitive, motor-cognitive, and executive function tasks.

	Correlation coefficient [95% Confidence interval]
Timed Up and Go (seconds)	−0.169 [−0.411–0.096]
Montreal Cognitive Assessment (/30)	0.143 [−0.120–0.387]
Body Position Spatial Task (product of score and span)	0.158 [−0.104–0.401]
Four Square Step Test (seconds)	−0.117 [−0.367–0.148]
Timed Up and Go-Cognitive (seconds)	−0.184 [−0.424–0.080]
Timed Up and Go percent time changec (%)	−0.010 [−0.270–0.251]
DKEFS Tower Test Total Achievement score (scaled) Total Achievement Scaled Score	−0.256 [−0.483–0.002]
DFKES Color Word Interference
Inhibition Scaled Score	−0.024 [−0.280–0.236]
Inhibition/Switching Scaled Score	0.192 [−0.088–0.414]
Inhibition Errors Scaled Score	0.152 [−0.111–0.395]
Inhibition/Switching Errors Scaled	0.171 [−0.091–0.411]
Trails B-A Difference score (seconds)	−0.118 [−0.365–0.145]
SF-12 Survey
Mental Health Composite Score	−0.302 [−0.520– −0.048]
Physical Health Composite Score	−0.204 [−0.439–0.057]

^c$\text{Formula:}\frac{\left(Tug-cog\right)-TUG}{TUG}*100\%$.

Both regression models showed participants with PH had significantly lower scores on the Physical Health Composite Scores and Mental Health Composite Scores of the SF-12 survey compared to individuals in the NPH group (Fig. 1, Table 3). Additionally, PH participants performed significantly worse on the DTT Total Achievement Score compared to NPH participants in both models (Fig. 2, Table 3).

Fig. 1.

Average SF-12 mental and physical health composite score for participants with and without a parental history of ADRD. Error bars represent standard error of the mean.

Table 3.

Performance on motor-cognitive and executive function tasks and SF-12 surveys between groups.

	Parental history, n = 34, Mean (SD) [range]	No parental history n = 24, Mean (SD) [range]	Model 1 a			Model 2 b
			β	p	R2	β	p	R2
Montreal Cognitive Assessment (/30)	25.6 (2.8) [18, 30]	24.7 (3.5) [14, 30]	0.59	0.491	0.053	0.59	0.499	0.057
Timed Up and Go (seconds)	8.2 (2.6) [5.6,20.9]	n = 23	−0.63	0.471	0.116	−0.73	0.403	0.170
		9.3 (4.0) [5.6,24.9]
Body Position Spatial Task (product score)	16.6 (9.9) [4, 49]	14.0 (4.6) [9, 20]	2.86	0.213	0.028	2.78	0.229	0.051
Four Square Step Test (seconds)	9.5 (2.0) [6,14.6]	n = 23	−0.40	0.600	0.047	−0.51	0.480	0.173
		10.2 (3.6) [6.3,23.5]
Timed Up and Go-Cognitive (seconds)	11.6(3.8) [6.6,22]	n = 23	−1.24	0.326	0.076	−1.43	0.227	0.22
		13.3 (5.5) [6.5,28.8]
Timed Up and Go costc (%)	43.6 (37.0) [−6.15,135.55]	n = 23	−2.86	0.771	0.015	−3.45	0.720	0.093
		44.4 (32.0) [0.2143.8]
DKEFS Tower of London Total Achievement score (scaled)	9.9 (2.0) [6,15]	11.5 (4.3) [6,23]	−1.78	0.046*	0.071	−1.78	0.050*	0.081
DFKES Color Word Interference
Inhibition Scaled Score	10.6 (2.4) [3,14]	10.7 (2.8) [5,15]	−0.19	0.795	0.003	−0.19	0.797	0.019
Inhibition/Switching Scaled Score	10.3(2.4) [4,14]	9.3 (3.2) [1,15]	0.86	0.281	0.039	0.86	0.290	0.045
Inhibition Errors Scaled Score	10.8 (2.2) [1,13]	10.0 (3.2) [1,13]	0.63	0.401	0.042	0.60	0.417	0.096
Inhibition/Switching Errors Scaled	10.5(2.3) [5,13]	9.5 (3.4) [1,14]	0.75	0.335	0.050	0.75	0.345	0.050
Trails B-A Difference score (seconds)	46.4 (26.3) [16.4109.3]	53.1 (31.0) [8.4125.3]	−4.62	0.557	0.033	−4.66	0.555	0.062
SF-12
Mental Health Composite Score	46.8(10.7) [2.5,59.4]	52.8(7.8) [36.3,62.6]	−7.17	0.007*	0.144	−7.14	0.008*	0.154
Physical Health Composite Score	40.9(9.3) [19.4,61.4]	44.7 (8.6) [28.9,60.5]	−5.60	0.023*	0.165	−5.60	0.025*	0.176

^aModel 1: Linear Regression adjusting for number of times leaving the house comparing Parental History and No Parental History Groups.

^bModel 2: Linear Regression adjusting for number of times leaving the house, body mass index, and hypertension comparing Parental History and No Parental History Groups.

^c$\text{Formula:}(\frac{\left(Tug-Cog\right)-TUG}{TUG})*100\%$.

^*P values indicate significant differences between Parental History and No Parental History Groups at the 0.05 level.

Fig. 2.

Average DKEFS Tower Test (DTT) total achievement scaled score for participants with or without a parental history of ADRD. Scores are scaled for age based on a normative group. Error bars represent standard deviation error of the mean.

4. Discussion and conclusion

The present, pilot study examined the relationship between PH of ADRD and performance on measures of executive function, motor-cognitive, and QOL tasks in African American women. Point biserial correlations indicate correlation of moderate effect size between family history and Mental and Physical Health Composite Score with those with a family history having worse scores. Weak or small effect sizes with no significant associations were seen for the other variables investigated. After adjusting for number of times participants left the house weekly, BMI, and hypertension, we observed differences between those with and without PH on the DTT task, a common test of planning and organization ability, in high functioning African American women and scored relatively high on the global screen with an average MoCA score of 25, compared to data from a large study which established a score of 22 as normative among African American adults [98]. Qualitative data on why MoCA scores were higher in this group than seen among other African American populations were not collected.

The participants with PH of ADRD performed significantly worse on the DTT task compared to NPH participants, which is consistent with previous studies. PH may be a risk factor for executive function decline and AD [65,66] in African American women in addition to other racial and gendered groups. The study expands on this literature by using a sample of African American women who are often not represented in research despite their increased risk for the development of AD compared with non-Hispanic Whites and men [27,81]. The authors acknowledge that caregiving can contribute to increased stress, which can influence cognitive decline as discussed previously. It is a limitation that more data are not available on the caregiver status of all participants and the psycho-social impact that caregiving had on the women in this sample who were caregivers.

Cognitive assessment with the DTT or similar tests that examine planning and organization may serve as an early predictor for future development of ADRD [99]. These individuals could benefit from longitudinal monitoring of further impairment and more regular follow-up and participate in appropriate rehabilitative interventions taking a multimodal approach. A recent study found that African American women with a high fat diet had increased serum TNF-alpha levels, which are critical in the pathology of AD [100] highlighting the importance of nutrition counselling in this high-risk group already showing initial stages of cognitive decline. Additionally, exercise and cognition training programs such as dance interventions have improved functioning, mental health, and cognition in a variety of populations [101–103]. Partnered dance exercise training can also reduce potential risk factors of ADRD including hypertension and obesity through regular activity [104]. We propose using a partnered dance exercise strategy along with nutrition counselling in our specific African American women population to help mitigate further decline and improve mental/physical health.

This pilot study had several limitations. The small sample size means that the study is underpowered to determine some effects. Possibly some of the youngest individuals in the NPH group have living parents that have not yet been diagnosed with ADRD but may be diagnosed in the future. However, those in the NPH group trended to being slightly older than those with PH. Although the focus in this study was on cognitive domains, e.g., executive function, that are preferentially affected early in an MCI process, or prior to diagnosis, future studies require the examination of memory measures, more visuospatial measures and language measures to more fully understand differences between African American women with PH and those without PH. The inability to consider the impact of caregiving on participants is also a limitation. Including caregiver status for all participants and the impact of caregiving on psycho-social health and controlling for caregiver status is important for future studies. Additionally, recruiting a more diverse sample of participants will be beneficial for generalizability as this sample included primarily middle to high income individuals in the Atlanta area. Overall, the findings should be considered useful for generating effect sizes that can power future studies that will more definitively provide conclusions.

In conclusion, with this pilot study we present initial evidence that a high-risk demographic of individuals with PH of ADRD may have reduced planning abilities and poorer physical and mental functioning. However, limitations in the study, including inability to analyze the impact that caregiving may have had on stress levels and psycho-social health among those individuals who were caregivers, necessitate further research. Further studies should investigate the relationship between PH of ADRD, executive function, and mental and physical health.

Abbreviations:

AD

Alzheimer's Disease

ADRD

Alzheimer's Disease and Related Dementias

APOE

Apolipoprotein E gene

PH

Parental History

QOL

Quality of Life

MCI

Mild Cognitive Impairment

IADLs

Instrumental Activities of Daily Living

NPH

No Parental History

BMI

Body Mass Index

MOCA

Montreal Cognitive Assessment

TUG

Timed Up and Go

TUG-cog

Timed Up and Go Cognitive

FSST

Four Square Step Test

BPST

Body Position Spatial Task

DTT

DKEFS Tower Test

SF-12

Short Form-12

Data availability statement

Data for this project will be made available upon request.