Alzheimer’s Disease and Related Dementia in Indigenous Populations: A Systematic Review of Risk Factors

Abstract

Background:

There remains a lack of information and understanding of the prevalence and incidence of Alzheimer’s disease and related dementia in Indigenous populations. Little evidence available suggests that Indigenous peoples may have disproportionately high rates of Alzheimer’s disease and related dementia (ADRD).

Objective:

Given this information, this study systematically explores what risk factors may be associated with ADRD in Indigenous populations.

Methods:

A search of all published literature was conducted in October 2016, March 2018, and July 2019 using Medline, Embase, and PsychINFO. Subject headings explored were inclusive of all terms related to Indigenous persons, dementia, and risk. All relevant words, phrases, and combinations were used. To be included in this systematic review, articles had to display an association of a risk factor and ADRD. Only studies that reported a quantifiable measure of risk, involved human subjects, and were published in English were included.

Results:

Of 237 articles originally identified through database searches, 45 were duplicates and 179 did not meet a priori inclusion criteria, resulting in 13 studies eligible for inclusion in this systematic review.

Conclusion:

Keywords

Alzheimer’s disease indigenous population neurocognitive disorders social determinants of health

INTRODUCTION

Dementia is characterized by multiple cognitive deficits, including impairment in memory [1]. Current estimates suggest that 46.8 million people worldwide are living with Alzheimer’s disease and related dementia (ADRD) [2]. In Canada specifically, the number of people with ADRD is projected to more than double over the next 25 years [3]. Based on data from the Framingham study, the lifetime risk of dementia at age 65 is 22% for women and 14% for men [4].

Little is currently known about the prevalence and incidence of ADRD among First Nations and Inuit populations, yet ADRD is expected to be an increasing challenge for home care services in First Nations and Inuit communities [5]. ADRD rates appear to be rising more rapidly in Indigenous than non-Indigenous populations [5]. Indigenous populations may already be experiencing disproportionately high ADRD rates [6, 7], as well as ADRD onset at younger ages [5, 6]. The current paucity of epidemiological information makes it difficult for communi-ties and service-providers to plan for, and respond to, this emerging issue. In particular, it is difficult to develop health promotion programs and policies without a fulsome understanding of the risk factors associated with ADRD in Indigenous populations [6].

There is no single cause of ADRD [8]. While genetic factors play an important role, less than 5% of Alzheimer’s disease patients have a clear diagnosis of familial Alzheimer’s disease [8]. In rare situations, there are some First Nations in Canada that exhibit specific familial Alzheimer’s disease [9]; however, ADRD more commonly arises from a combination of genetic and environmental risk factors. While increasing age is the most predictive factor in assessing risk for ADRD, research into other risk and protective factors for ADRD has helped to shape an understanding of overall ADRD risk [10]. Non-modifiable, or biological, factors associated with dementia include old age and female sex [11]. Potentially modifiable, or social, factors that may contribute to risk include low education and low socioeconomic status [11 –13]. Other potentially modifiable factors include alcohol abuse, smoking, and reduced physical activity [11 , 15]. Medical risk factors, such as high blood pressure, high cholesterol, being overweight, diabetes, and cardiovascular diseases, may also be contributing factors [16]. It is estimated that one-third of all Alzheimer’s disease cases globally can be attributed to potentially modifiable risk factors [11, 17].

Despite the paucity of research in Indigenous specific risk and protective factors for ADRD, there has been a surge in information on risk factors for the general population, including the identification of childhood education, exercise, maintaining social engagement, reducing smoking, and management of hearing loss, depression, diabetes, and obesity as potential factors to delay or even prevent certain dementia cases [18].

A life-course approach is helpful in understanding disease risk factors in Indigenous populations from a holistic perspective [19, 20]. In relation to dementia, this approach can account for the impact of life events on neuro-cognitive growth and decline [21]. By assessing potentially modifiable risk factors from different phases of the lifespan, we may better decrease the future incidence in addition to successful elimination of the most potent factors [18]. Our analysis sought to delineate the factors associated with ADRD risk into those that occur in early, mid and later life in order to highlight potential targets for interventions.

Despite great diversity of Indigenous peoples, many similarities remain with regards to the determinants of their health and illnesses [22]. Consequently, the results from this study will be framed using an Indigenous perspective of the social determinants of health [23]. This includes moving beyond a deficit-focused perspective and identifying potential factors that may help protect against ADRD [24], as well as by adopting a holistic life-course approach [20], to the extent allowed by available literature. Given the growing evidence that Indigenous peoples have disproportionately high rates of ADRD [6, 7], this study systematically explores what risk factors may be associated with ADRD in Indigenous populations.

METHODS

Throughout this manuscript, we followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement to report this systematic review.

Inclusion criteria

To be included in this systematic review, articles had to display an association of a risk factor and ADRD. Only studies that reported a quantifiable measure of risk, involved human subjects, and were published in English were included. Conference proceedings, books, editorial publications, letters, and commentaries were not included. Articles were not excluded based on study design.

Search strategy

A search of all published literature was conducted in October 2016, March 2018, and July 2019 using Medline, Embase, and PsychINFO. Subject headings explored were inclusive of all terms related to Indigenous persons, dementia, and risk. All relevant words, phrases, and combinations were used. All possible synonyms, alternate terminologies, variant word endings were also adopted. Boolean logic was used to combine concepts and drop irrelevant articles. The complete search was as follows: (risk OR “risk factors”) AND (Alzheimer* OR Dementia OR memory) AND (Aboriginal OR “First Nation*” OR Indigenous OR Inuit OR Metis).

Study selection

We undertook a systematic, critical review. Articles were stored in Endnote X8 [25] and duplicates were removed. After all duplicates were removed, abstracts of remaining records were screened according to the eligibility criteria. Titles, abstracts, and papers were independently assessed by two reviewers (GS and KL). Differences of opinions were resolved by consensus.

Data extraction

We extracted the characteristics of each included study, including: author, study design, risk factors, measure of association, study groups, and main results. If articles did not provide enough information to extract relevant data, authors were contacted in an attempt to acquire additional information. A second reviewer checked all abstracted data for completeness and accuracy.

Quality assessment

The methodological quality of this systematic re-view was assessed using the Assessing the Meth-odologic Quality of Systematic Reviews tool (AMSTAR) [26]. Individual studies were assessed based on a predetermined set of criteria. There were two distinct quality assessment checklists developed for observational studies and review studies (see the Supplementary Material). The two quality assessment checklists were developed and adapted from the National Institute of Health (NIH) Study Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies and the NIH Study Quality Assessment Tool for Systematic Reviews and Meta-Analyses [27]. The original checklists were modified through a consensus building process among reviewers, namely, modifications to include strength-based indicators, First Nations community engagement, and culturally sensitive methods and reporting. Strengths-based indicators have commonly been used in fields such as social work, psychology, and international development, and their constructive, holistic, and community-based focus makes them useful tools in Indigenous health research [28, 29]. According to the consolidated criteria for strengthening reporting of health research involving Indigenous peoples (the CONSIDER statement), the inclusion of strengths-based indicators helps reduce harmful stereotyping [28]. The CONSIDER statement also recommends conducting Indigenous health research in collaboration with the relevant communities and, in this vein, we chose to include community engagement as an item in our quality assessment checklist [28].

RESULTS

Of 237 articles originally identified through data-base searches, 45 were duplicates and 179 did not meet a priori inclusion criteria, resulting in 13 studies eligible for inclusion in this systematic review (Fig. 1).

Fig. 1

Of the studies meeting inclusion criteria, ten are observational studies and three are review articles. Of the ten observational studies included, six were cross-sectional studies and four were cohort studies; of the cohort studies, two were prospective and one was longitudinal. Odds ratios were the most frequent means of assessing the relationship between ADRD and potential risk factors (7/10 studies). Half of the observational studies (5/10) examined ADRD risk factors in Aboriginal Australian populations. Indigenous populations in Canada and the Chamorros of Guam were each the subject of two observational studies, and one study was about the Melanau people of East Malaysia (Table 1).

Table 1

Risk factors explored/identified for ADRD in Indigenous populations (both significant and non-significant)

No.	Title	Author, Year	Journal	Study population	Study design	Measure of association	Risk factor(s) (significance)	Risk factor(s) (No significance)
1	Factors associated with the high prevalence of dementia in older Aboriginal Australians	Radford, et al., 2018 [32]	Journal of Alzheimer’s Disease	Aboriginal and Torres Strait Islander people in Australia	Cross-sectional	Odds ratio	Age (1.96, 95% CI: 1.43–2.68), childhood trauma (1.63, 1.11–2.39), high-risk alcohol use (2.6, 1.09–6.21), unskilled job (2.25, 1.14–4.44), stroke (4.36, 2.16–8.77), head injury with loss of consciousness (2.87, 1.44–5.74), and epilepsy (4.65, 1.62–13.34) were associated with dementia.	Male sex (1.66, 95% CI: 0.85–3.24), urban site (0.98, 0.49–1.93), years of education (0.82, 0.59–1.15), grew up in inner regional area (1.276, 0.55–2.95) and outer regional/remote area (1.308, 0.57–3.02), socioeconomic advantage (1.39, 1.00–1.94), poor/fair child health (0.95, 0.41–2.22), death of parent(s) (0.89, 0.39–2.02), Retrospective Indigenous Childhood Enrichment Scale (RICE) (0.87, 0.61–1.24), smoking (1.14, 0.80–1.64), low risk alcohol consumption (1.01, 0.38–2.73), culture is source of strength (0.514, 0.24–1.08), connected to community (0.61, 0.19–1.95), police custody (1.87, 0.92–3.82), depression (1.79, 0.90–3.58), anxiety/PTSD (0.96, 0.44–2.08), Diabetes (1.66, 0.85–3.24), hypercholesterolemia (1.03, 0.52–2.02), hypertension (0.72, 0.36–1.45), heart disease (0.63, 0.32–1.25), hearing loss (1.98, 0.99–3.96)
2	Childhood Stress and Adversity is Associated with Late-Life Dementia in Aboriginal Australians	Radford et al., 2017 [31]	American Journal of Geriatric Psychiatry	Aboriginal and Torres Strait Islander people in Australia	Cross-Sectional	Odds ratio	Each SD increase to childhood adversity score was associated with all-cause dementia (1.70, 95% CI: 1.14–2.54) and Alzheimer dementia (1.77, 1.08–2.91)
3	Implications of Risk Factors for Alzheimer’s Disease in Canada’s Indigenous Population	MacDonald et al., 2015 [37]	Canadian Geriatrics Journal	Indigenous people living on- and off- reserve and non-Indigenous people in Canada	Cohort	Population attributable risk (PAR)	Diabetes mellitus (6.0%), midlife hypertension (14.2%), midlife obesity (16.8%), physical inactivity (32.5%), smoking (19.4%), and low educational attainment (232.4%) contribute to dementia rates in Indigenous population.	All differences in PAR are significant at p < 0.001
4	Incidence and predictors of cognitive impairment and dementia in Aboriginal Australians: A follow-up study of 5 years	Lo Giudice et al., 2015 [46]	Alzheimer’s & Dementia	Aboriginal people in Australia	Longitudinal Cohort	Odds ratio	Age (1.12, 95% CI: 1.06–1.17), head injury (5.22, 1.85–14.70), analgesic medication (3.60, 1.35–9.62), poor mobility (3.08, 1.09, 8.72) and BMI (0.90, 0.82–0.98) were associated with dementia.	Non-significant variables removed using backward stepwise approach. Male sex (0.88, 0.45–1.71), drink alcohol (0.75, 0.33–1.68), drink every day (0.65, 0.07–5.76), 0.53, 0.17–1.63), current smoker (0.45, 0.18–1.10), chew tobacco (1.65, 0.84–3.22), poor vision (1.17, 0.60–2.29), poor hearing (1.50, 0.68–3.30), pain always present (0.37, 0.12–1.11), fell in last year (1.71, 0.86–3.40), diabetes (1.02, 0.52–2.03), hypertension (1.96, 0.84–4.59), heart problem (0.83, 0.40–1.71), kidney problem (1.29, 0.63–2.68), poor kidney function (1.52, 0.73–3.16), head injury (1.87, 0.95–3.69), antihypertensive medication (0.87, 0.42–1.82), lipid-lowering medication (0.61, 0.31–1.20), glucose-lowering medication (0.54, 0.28–1.10), antithrombotic medication (0.65, 0.33–1.29), HbA1c measure of blood sugar (0.88, 0.67–1.15), systolic blood pressure (mmHg) (1.02, 1.00–1.03), diastolic blood pressure (1.00, 0.98–1.03)
5	Dementia among elderly Melanau: A community survey of an indigenous people in East Malaysia	Inu Pu’un, et al., 2014 [33]	International Medical Journal	Melanau people	Cross-sectional	Odds ratio	Age (1.19, 95% CI: 1.12–1.27), no education (7.56, 1.70–33.57) and multiple cardiovascular illnesses (3.76, 1.25–11.28) were associated with dementia.	Female gender (1.40, 95% CI: 0.68–2.91), never married (1.85, 0.39–8.89), unskilled employment (3.17, 0.73–13.77), no employment (1.57, 0.28–8.98), 1–2 cardiovascular illnesses (1.30, 0.58–2.93), family history of dementia (1.04, 0.43–2.49), smoking (0.72, 0.29–1.80)
6	The emergence of dementia as a health concern among First Nations populations in Alberta, Canada	Jacklin et al., 2012 [6]	Canadian Journal of Public Health	First Nations and Non-First Nations populations in Canada	Prospective Cohort	Prevalence	First Nations males were statistically more likely to be diagnosed with dementia relative to non-First Nations males (p = 0.017).	No other risk factors investigated.
7	Factors associated with dementia in Aboriginal Australians	Smith et al., 2010 [34]	Australian New Zealand Journal of Psychiatry	Aboriginal people in Australia	Cross-sectional	Odds ratio	Age 80+ (40.3, 95% CI: 10.3–156.8), male gender (3.1, 1.4–6.8), no education (2.7, 1.1–6.7), smoking (4.5, 1.1–18.6); stroke (17.2, 5.9–49.7), epilepsy (33.5, 4.8–232.3), head injury (4.0, 1.7–9.4), falls (2.7, 1.2–6.1), poor mobility (13.3, 4.1–43.9), daytime urinary incontinence (116.8, 21.9–622.8), nighttime urinary incontinence (87.4, 18.4–415.7), urine problems (4.2, 1.8–10.2) were associated with dementia.	Drinks alcohol (1.8, 95% CI: 0.5–6.5), smoker when young (0.5, 0.20–1.0), chews tobacco (0.5, 0.2–1.2, diabetes (0.9, 0.4–1.9), hypertension (0.8, 0.4–1.6), heart problems (1.4, 0.6–3.4), poor vision (0.8, 0.4–1.8), poor hearing (0.8, 0.3–2.2), pain (0.6, 0.2–1.4)
8	High prevalence of dementia and cognitive impairment in Indigenous Australians	Smith et al., 2008 [45]	Neurology	Aboriginal people Australia	Cross-sectional	Prevalence	Aboriginal adults aged 70+ (p < 0.0001) were more frequently diagnosed with dementia relative to non-Aboriginal adults aged 70+. Male Aboriginal Australians were more frequently diagnosed with dementia relative to female Aboriginal Australians (17% vs 9%, 95% CI of the difference: 0.4% –15.6%).
9	Prevalence of dementia in Chamorros on Guam: relationship to age, gender, education, and APOE	Galasko et al., 2007 [38]	Neurology	Chamorros	Cross-sectional	Odds ratio	Age (1.11, 1.08–1.13) and low education were associated with dementia.	Gender (0.97, 0.71–1.32)
10	Two sites in the MAPT region confer genetic risk for Guam ALS/PDC and dementia	Sundar et al., 2007 [47]	Human Molecular Genetics	Chamorros	Prospective cohort	Odds ratio	Microtubule-associated protein tau gene (MAPT), site 2 (1.61, 95% CI: 1.0–2.62), site 9 (1.06, 0.57–1.97), site 6 and 9 (3.02, 1.10–8.25)	All sites significant at p≤0.05.

Of the three review articles included, two are systematic reviews that explore global prevalence of ADRD in Indigenous populations as well as potential risk factors. The third article is a literature review that explores the particular context of ADRD among Indigenous peoples in the southwestern United States (Table 2).

Table 2

Review articles reporting risk factors for ADRD in Indigenous populations

No.	Title	Author, Year	Journal	Study design	Inclusion criteria	Risk factor(s)
1	Dementia and Cognitive Impairment Prevalence and Associated Factors in Indigenous Populations	de Souza-Talarico et al., 2016 [35]	Alzheimer Disease and Associated Disorders	Systematic review	1) Elderly Indigenous population; 2) Cognitive impairment /dementia descriptors; and 3) Prevalence statistic(s) reported	Increased age, decreased education level, poor mobility, head injury, analgesic medications, low body mass index, stroke, hypertension
2	Prevalence and incidence of dementia among indigenous populations: a systematic review	Warren et al., 2015 [7]	International Psychogeriatrics	Systematic review	1) Indigenous population; 2) Cognitive impairment typically occurred over age 45; and 3) Prevalence or incidence statistics reported	Lower education, socioeconomic status, higher rates of hypertension, cardiovascular disease and alcohol abuse, aging population structure and poorer overall health.
3	Challenges to the recognition and assessment of Alzheimer’s disease in American Indians of the southwestern United States	Griffin-Pierce et al., 2008 [36]	Alzheimer’s & Dementia	Literature review	Not specified	Higher risk of stroke

Quality assessment

Of the ten observational studies included, all but one (9/10) had a clearly defined research question. All studies outlined and defined a study population to include an Indigenous cohort and defined valid and reliable outcome measures. Most reported age-/sex-adjusted results (6/10) and identified confounding variables (7/10). However, only three of the studies modelled positive outcomes and strength-based factors. Half (5/10) described any type of community engagement (Table 3).

Table 3

Quality Assessment Checklist Results: Observational Studies

Question	Radford et al., 2018 [32]	Radford et al., 2017 [31]	MacDonald et al., 2015 [37]	Lo Giudice et al., 2015 [46]	Inu Pu’un, et al., 2014 [33]	Jacklin et al., 2012 [5]	Smith et al., 2010 [34]	Smith et al., 2008 [45]	Galasko et al., 2007 [38]	Sundar et al., 2007 [47]
Was the research question of objective clearly stated? (Y/N)	Y	Y	Y	Y	Y	Y	Y	Y	Y	N
Was the study population clearly specified and defined? (i.e., Indigenous groups included) (Y/N)	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
Were the outcome measures clearly defined, valid, reliable and implemented consistently across all participants (Y/N)	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
Was there engagement or co-authorship with Indigenous communities/people/organizations? (Y/N)	Y	Y	N	Y	N	Y	Y	Y	N	N
Were positive outcomes and strength-based factors modeled? (i.e. not deficit-based) (Y/N)	Y	Y	N	N	Y	N	N	N	N	N
Do the data support the authors’ interpretation? (Y/N)	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
Were the results age and/or sex adjusted? (Y/N)	Y	Y	N	Y	Y	Y	Y	Y	Y	N
Were confounders identified? (Y/N)	Y	Y	N	Y	Y	N	Y	Y	Y	N
Can causality be assumed? (Y/N)	N	N	N	N	N	N	N	N	N	N

Three review articles were included in this systematic review. Two of the studies had clearly focused research questions, used appropriate inclusion criteria, and assessed the level of evidence of the primary articles, whereas the third review article did not meet any of these criteria. All three articles had data to support their interpretations (Table 4).

Table 4

Quality Assessment Checklist: Review Articles

Question	de Souza-Talarico et al., 2016 [35]	Warren et al., 2015 [7]	Griffin-Pierce et al., 2008 [36]
Did the authors have a clearly focused research question? (Y/N)	Y	Y	N
Were appropriate inclusion criteria used to select articles? (Y/N)	Y	Y	N
Was the quality of the primary articles included assessed? (Y/N)	Y	Y	N
Do the data support the authors’ interpretation? (Y/N)	Y	Y	Y

A quality assessment of our entire review was undertaken. The methodological quality of this systematic review was assessed using the Assessing the Methodologic Quality of Systematic Reviews tool (AMSTAR) [26]. Using this previously validated framework, our systematic review received a ‘moderate’ quality score. A moderate ranking reveals the systematic review has more than one weakness but no critical flaws; it may provide an accurate summary of the results of the available studies that were included in the review [30]. The AMSTAR tool was not specifically developed as a quality assessment for Indigenous health studies. Some of the nuances in the checklist may not be suitable for this specific review and may have artificially deflated the quality score.

Risk factors

Extracted risk factors for ADRD were sorted by reviewers into potentially modifiable and non-modifiable factors. In total, 14 potentially modifiable risk factors were identified; of those, cardiovascular disease and low educational attainment were the most commonly reported, with seven referencing articles apiece. Head injury was identified as risk factor in four of the studies included, while hypertension and poor mobility were reported in three, and childhood trauma, smoking, low body mass index, and alcohol use were referenced by two articles apiece. All remaining potentially modifiable risk factors were reported in only one article (Tables 1 and 5).

Table 5

Potentially modifiable and non-modifiable risk factors for ADRD in Indigenous populations (only those with statistical significance)

	Risk Factors	Observational Studies	Review Studies
Potentially Modifiable	Cardiovascular disease	Radford et al., 2018 [32]	de Souza-Talarico et al., 2016 [35]
		Inu Pu’un et al., 2014 [33]	Warren et al., 2015 [7]
		Smith et al., 2010 [34]	Griffin-Pierce et al., 2008 [36]
	Low educational attainment	MacDonald et al., 2015 [37]	de Souza-Talarico et al., 2016 [35]
		Inu Pu’un et al., 2014 [33]	Warren et al., 2015 [7]
		Galasko et al., 2007 [38]
		Smith et al., 2010 [34]
	Head injury	Radford et al., 2018 [32]	de Souza-Talarico et al., 2016 [35]
		Lo Giudice et al., 2015 [46]
		Smith et al., 2010 [34]
	Hypertension	MacDonald et al., 2015 [37]	de Souza-Talarico et al., 2016 [35]
		Warren et al., 2015 [7]
	Poor mobility	Smith et al., 2010 [34]	de Souza-Talarico et al., 2016 [35]
		Lo Giudice et al., 2015 [46]
	Alcohol use	Radford et al., 2018 [32]	Warren et al., 2015 [7]
	Childhood trauma	Radford et al., 2018 [32]
		Radford et al., 2017 [31]
	Smoking	MacDonald et al., 2015 [37]
		Smith et al., 2010 [34]
	Low body mass index	Lo Giudice et al., 2015 [46]	de Souza-Talarico et al., 2016 [35]
	Socioeconomic status		Warren et al., 2015 [7]
	Unskilled job	Radford et al., 2018 [32]
	Obesity	MacDonald et al., 2015 [37]
	Physical inactivity	MacDonald et al., 2015 [37]
	Diabetes mellitus	MacDonald et al., 2015 [37]
Non-Modifiable	Increasing age	Radford et al., 2018 [32]	de Souza-Talarico et al., 2016 [35]
	Lo Giudice et al., 2015 [46]	Warren et al., 2015 [7]
		Inu Pu’un et al., 2014 [33]
		Smith et al., 2010 [34]
		Smith et al., 2008 [45]
		Galasko et al., 2007 [38]
	Male gender	Jacklin et al., 2012 [5]
		Smith et al., 2010 [34]
		Smith et al., 2008 [45]
	Epilepsy	Radford et al., 2018 [32]
		Smith et al., 2010 [34]
	Genetics	Sundar et al., 2007 [47]

Fewer non-modifiable risk factors were identified. Increasing age was the most commonly reported; it was referenced in eight of the 13 articles reviewed. Male gender was reported as a risk factor in three studies, epilepsy was reported in two, and genetic polymorphisms were reported in one (Tables 1 and 5). It is important to acknowledge that the above listed risk factor counts are inclusive of all 13 articles in this review. With three review articles included, there is some overlap in primary sources. A complete list of statistically significant potentially modifiable and non-modifiable risk factors and their sources can be found in Table 5.

Potential protective factors

Four observational studies, in addition to reporting risk factors, also suggested factors that may protect against ADRD in Indigenous populations. Radford and colleagues [31, 32] observed that older Aboriginal Australians who reported being connected to their culture had a non-statistically significant reduction in ADRD risk, and hypothesize that cultural continuity may support healthy aging in this population. Radford and colleagues [31] also observed an inverse relationship between family size and childhood adversity score, and posit that support from extended family may help mitigate experiences of childhood trauma. Similarly, Pu’un, Othman, and Drahman [33] found no association between employment, gender, or marital status with dementia; in their opinion, this suggests that supportive environmental factors, such as close-knit familial and social connections, may be protective against ADRD. Jacklin, Walker, and Shawande [6] observe that health promotion programs tailored to community needs may be an effective way to mitigate some of the risk factors associated with ADRD.

Negative findings

Beyond the non-significant potentially-protective findings identified above, several other risk factors were explored and found to have no association/significance with ADRD (Table 1). When comparing the observational studies included in this review, there were conflicting findings on several potentially modifiable and non-modifiable risk factors, such that certain studies found significance while other studies did not. Male sex, smoking, hypertension, cardiovascular disease, alcohol consumption, head injury, educational attainment, and diabetes were found to be statistically significant risk factors in certain articles (Table 5) and non-significant risk factors in others (Table 1). Furthermore, place of residence, childhood health, childhood enrichment, death of parents, police custody, depression, anxiety/PTSD, hypercholesterolemia, hearing loss, chewing tobacco use, poor vision, poor hearing, chronic pain, fell in last year, and kidney problems were explored and found to be non-significant risk factors (Table 1).

DISCUSSION

Potentially modifiable risk factors for ADRD in Indigenous populations are distributed throughout the life cycle (Fig. 2). The two most frequently reported potentially modifiable risk factors, low educational attainment and cardiovascular disease, are especially relevant during childhood/young adulthood and adulthood/older age, respectively [7 , 33–38]. A number of ADRD risk factors that manifest mid-life, such as obesity, hypertension, and diabetes mellitus, may themselves be related to conditions much earlier in life. For example, certain intrauterine factors, like maternal obesity or gestational diabetes, may predispose children to developing metabolic syndrome [39 –41]. Both indirect and direct risk factors for ADRD can occur early in life, illustrating the importance of using a life-course approach to better understand ADRD in Indigenous populations.

Fig. 2

Potentially modifiable risk factors for Alzheimer’s disease and related dementia in Indigenous populations span the lifespan.

The large number of potentially modifiable risk factors reported relative to non-modifiable risk factors illustrates the importance of socioeconomic context in the pathogenesis of ADRD in Indigenous populations. There can be a tendency to prioritize genetic over social explanations when encountering disproportionately high disease rates in Indigenous populations [42]. Unfortunately, this can distract from modifiable proximal, intermediate, and distal determinants of health [19, 42]. One study from Canada estimates that 79.0% of ADRD cases among First Nations people living on-reserve can be attributed to educational and cardiovascular risk factors, compared to 61.7% of cases in the general Canadian population, a statistically significant difference [37]. Cases of ADRD in Indigenous populations in Canada are disproportionately caused by potentially modifiable factors.

There appear to be other ADRD risk factor differences between Indigenous and non-Indigenous populations, and some of these differences may be explained by how various risk factors are distributed in Indigenous relative to non-Indigenous populations. For example, Indigenous populations in Canada and Australia are understood to have disproportionately high rates of cardiovascular disease [22 , 44], and this may explain why cardiovascular disease was identified as a key risk factor for ADRD in Indigenous populations alongside low educational attainment [7 , 32–38]. By contrast, in non-Indigenous populations, low educational attainment stands out as the predominant potentially modifiable ADRD risk factor [11].

Another difference in ADRD risk factors for Indigenous relative to non-Indigenous populations relates to sex. In non-Indigenous populations, female sex is associated with increased risk of ADRD [11]. By contrast, several studies reviewed for this systematic review suggest an association between male sex and ADRD risk in Indigenous populations [5 , 45]. The reasons for this are not yet understood [5].

In addition to risk factors, our investigation also explored factors that may protect against ADRD. Of the ten observational studies included in this systematic review, only four discussed potential protective factors. Given what appears to be a disproportionately high burden of potentially modifiable risk ADRD risk factors in Indigenous populations, more attention should be directed towards ameliorative factors, programs, policies, and initiatives, and how they can be promoted within Indigenous communities.

Due to the heterogeneity of measurements and study designs, a meta-analysis of risk factors was not possible. The studies meeting inclusion criteria for this review provide incomplete coverage of global Indigenous populations. Half of the included observational articles described risk factors in Indigenous populations in Australia [31 , 46] and the remaining five focused on populations in Canada [5, 37], Guam [38, 47], and East Malaysia [33]. No observational studies describing ADRD risk factors in Indigenous populations in the United States were found. Although Indigenous populations around the world tend to face similar health-related challenges [22], there is also considerable diversity with regards to Indigenous cultures and experiences of colonialism [48], raising some concerns about the external validity of the patterns described in this systematic review.

As a result of the paucity of data on this subject, there was some overlap in the primary data reported. That is, the two review articles included data reported in some of the other articles included in this systematic review. Even amongst the non-review articles, there was some duplicity in data reported. Despite this, the decision was made to include all articles that met our inclusion criteria with the mindset that each provided valuable and distinct perspectives and information.

Several factors—namely, urinary incontinence [34], analgesic use [35, 46], and falls [34]—were occasionally reported as factors associated with ADRD risk. All three of these conditions can be caused, and are exacerbated, by ADRD [49 –53]. They are most likely co-morbid conditions instead of risk factors, and so they were excluded from the list of risk factors identified in this systematic review. A more nuanced differentiation between ADRD risk factors and co-morbidities in Indigenous populations is a potential topic for further research.

The results of this study, namely the lack of arti-cles included in this review echoes the need for further research on the risk factors associated with ADRD in culturally, ethnically, and socioeconomically diverse populations including Indigenous populations as identified by Radford and colleagues [32]. Future research is urgently needed to distinguish the genetic and social risk factors associated ADRD in Indigenous populations as it continues to grow in incidence. A continued focus on the life-course social determinants of health would appear to be a crucial factor in addressing the burden of this disease in Indigenous communities.

Footnotes

ACKNOWLEDGMENTS

We are grateful to the work of all research teams whose work contributed to this systematic review.

Authors’ disclosures available online ().

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

References

American Psychiatric Association (2013) Diagnostic and Statistical Manual of Mental Disorders. American Psychiatric Association, Arlington.

Prince

, Wimo

, Guerchet

, Ali

G-C

, Wu

Y-T

, Prina

(2015) World Alzheimer Report 2015: The global impact of dementia: An analysis of prevalence, incidence, cost and trends. Alzheimer’s Disease International, London.

Ward-Griffin

, Hall

, DeForge

, St-Amant

, McWilliam

, Oudshoorn

, Forbes

, Klosek

(2012) Dementia home care resources: How are we managing?. J Aging Res 2012, 590724.

Seshadri

, Beiser

, Kelly-Hayes

, Kase

, Au

, Kannel

, Wolf

(2006) The lifetime risk of stroke: Estimates from the Framingham Study. Stroke 37, 345–350.

Jacklin

, Walker

(2012) Trends in Alzheimer’s disease and related dementias among First Nations and Inuit. Unpublished document Prepared for Health Canada, First Nations and Inuit Health, Home and Community Care Program. Government of Canada, Ottawa.

Jacklin

, Walker

, Shawande

(2013) The emergence of dementia as a health concern among First Nations populations in Alberta, Canada. Can J Pub Health 104, e39–e44.

Warren

, Shi

, Young

, Borenstein

, Martiniuk

(2015) Prevalence and incidence of dementia among indigenous populations: A systematic review. Int Psychogeriatr 27, 1959–1970.

U.S. National Library of Medicine, Alzheimer Disease, https://ghr.nlm.nih.gov/condition/alzheimer-disease

Butler

, Dwosh

, Beattie

, Guimond

, Lombera

, Brief

, Illes

, Sadovnick

(2011) Genetic counseling for early-onset familial Alzheimer disease in large aboriginal kindred from a remote community in British Columbia: Unique challenges and possible solutions. J Genet Counsel 20, 136–142.

10.

Diamond

(2006) A report on Alzheimer’s disease and current research, Alzheimer Society of Canada.

11.

Eratne

, Loi

, Farrand

, Kelso

, Velakoulis

, Looi

(2018) Alzheimer’s disease: Clinical update on epidemiology, pathophysiology and diagnosis. Australas Psychiatry 26, 347–357.

12.

Goldbourt

, Schnaider-Beeri

, Davidson

(2007) Socioeconomic status in relationship to death of vascular disease and late-life dementia. J Neurol Sci 257, 177–181.

13.

Scazufca

, Menezes

, Vallada

, Crepaldi

, Pastor-Valero

, Coutinho

, Di Rienzo

, Almeida

(2008) High prevalence of dementia among older adults from poor socioeconomic backgrounds in Sao Paulo, Brazil. Int Psychogeriatr 20, 394–405.

14.

King

(1986) Alcohol abuse and dementia. Int J Geriatr Psychiatry 1, 31–36.

15.

Ridley

, Draper

, Withall

(2013) Alcohol-related dementia: An update of the evidence. Alzheimers Res Therapy 5, 3.

16.

Williams

, Plassman

, Burke

, Holsinger

, Benjamin

(2010) Preventing Alzheimer’s disease and cognitive decline. Evid Rep Technol Assess (Full Rep) 193, 1–727.

17.

Norton

, Matthews

, Barnes

, Yaffe

, Brayne

(2014) Potential for primary prevention of Alzheimer’s disease: An analysis of population-based data. Lancet Neurol 13, 788–794.

18.

Livingston

, Sommerlad

, Orgeta

, Costafreda

, Huntley

, Ames

, Ballard

, Banerjee

, Burns

, Cohen-Mansfield

(2017) Dementia prevention, intervention, and care. Lancet 390, 2673–2734.

19.

Reading

, Wien

(2009) Health inequalities and social determinants of Aboriginal peoples’ health, National Collaborating Centre for Aboriginal Health, Prince George, BC.

20.

Priest

, Mackean

, Waters

, Davis

, Riggs

(2009) Indigenous child health research: A critical analysis of Australian studies. Aust N Z J Public Health 33, 55–63.

21.

Arkles

, Jackson Pulver

, Robertson

, Draper

, Chalkley

, Broe

(2010) Ageing, cognition and dementia in Australian Aboriginal and Torres Strait Islander peoples: A life cycle approach. Neuroscience Research Australia and University of New South Wales, Sydney.

22.

Gracey

, King

(2009) Indigenous health part 1: Determinants and disease patterns. Lancet 374, 65–75.

23.

Greenwood

, De Leeuw

, Lindsay

, Reading

(2015) Determinants of Indigenous Peoples’ Health, Canadian Scholars’ Press.

24.

Lind

, Smith

(2008) Analyzing the state of community health nursing: Advancing from deficit to strengths-based practice using appreciative inquiry. Adv Nurs Sci 31, 28–41.

25.

Clarivate Analytics (2016) EndNote X8. Philadelphia.

26.

Shea

, Grimshaw

, Wells

, Boers

, Andersson

, Hamel

, Porter

, Tugwell

, Moher

, Bouter

(2007) Development of AMSTAR: A measurement tool to assess the methodological quality of systematic reviews. BMC Med Res Methodol 7, 10.

27.

National Institute ofHealthNationalHeart, Lung and Blood Institute Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies. https://www.nhlbi.nih.gov/health-pro/guidelines/in-develop/cardiovascular-risk-reduction/tools/cohort

28.

Huria

, Palmer

, Pitama

, Beckert

, Lacey

, Ewen

, Smith

(2019) Consolidated criteria for strengthening reporting of health research involving indigenous peoples: The CONSIDER statement. BMC Medical Res Methodol 19, 173.

29.

Rountree

, Smith

(2016) Strength-based well-being indicators for Indigenous children and families: A literature review of Indigenous communities’ identified well-being indicators. Am Indian Alsk Native Ment Health Res 23, 206–220.

30.

Shea

, Reeves

, Wells

, Thuku

, Hamel

, Moran

, Moher

, Tugwell

, Welch

, Kristjansson

(2017) AMSTAR 2: A critical appraisal tool for systematic reviews that include randomised or non-randomised studies of healthcare interventions, or both. BMJ 358, j4008.

31.

Radford

, Delbaere

, Draper

, Mack

, Daylight

, Cumming

, Chalkley

, Minogue

, Broe

(2017) Childhood stress and adversity is associated with late-life dementia in Aboriginal Australians. Am J Geriatric Psychiatry 25, 1097–1106.

32.

Radford

, Lavrencic

, Delbaere

, Draper

, Cumming

, Daylight

, Mack

, Chalkley

, Bennett

, Garvey

(2019) Factors associated with the high prevalence of dementia in older Aboriginal Australians. J Alzheimers Dis 70, S75–S85.

33.

Pu’un

, Othman

, Drahman

(2014) Dementia among elderly Melanau: A community survey of an indigenous people in East Malaysia. Int Med J 21, 1–4.

34.

Smith

, Flicker

, Dwyer

, Atkinson

, Almeida

, Lautenschlager

, LoGiudice

(2010) Factors associated with dementia in Aboriginal Australians. Aust N Z J Psychiatry 44, 888–893.

35.

de Souza-Talarico

, de Carvalho

, Brucki

, Nitrini

, Ferretti-Rebustini

(2016) Dementia and cognitive impairment prevalence and associated factors in indigenous populations. Alzheimer Dis Assoc Disord 30, 281–287.

36.

Griffin-Pierce

, Silverberg

, Connor

, Jim

, Peters

, Kaszniak

, Sabbagh

(2008) Challenges to the recognition and assessment of Alzheimer’s disease in American Indians of the southwestern United States. Alzheimers Dement 4, 291–299.

37.

MacDonald

, Barnes

, Middleton

(2015) Implications of risk factors for Alzheimer’s disease in Canada’s Indigenous population. Can Geriatr J 18, 152–158.

38.

Galasko

, Salmon

, Gamst

, Olichney

, Thal

, Silbert

, Kaye

, Brooks

, Adonay

, Craig

U-K

(2007) Prevalence of dementia in Chamorros on Guam: Relationship to age, gender, education, and APOE. Neurology 68, 1772–1781.

39.

Boney

, Verma

, Tucker

, Vohr

(2005) Metabolic syndrome in childhood: Association with birth weight, maternal obesity, and gestational diabetes mellitus. Pediatrics 115, e290–296.

40.

Junien

, Nathanielsz

(2007) Report on the IASO Stock Conference 2006: Early and lifelong environmental epigenomic programming of metabolic syndrome, obesity and type II diabetes. Obes Rev 8, 487–502.

41.

Meigs

, Wilson

, Fox

, Vasan

, Nathan

, Sullivan

, D’Agostino

(2006) Body mass index, metabolic syndrome, and risk of type 2 diabetes or cardiovascular disease. J Clin Endocrinol Metab 91, 2906–2912.

42.

Polanco

, Arbour

(2018) Type 2 diabetes in Indigenous populations: Why a focus on genetic susceptibility is not enough. In Determinants of Indigenous Peoples’ Health: Beyond the Social, Greenwood M, De Leeuw S, Lindsay NM, eds. Canadian Scholars’ Press, Toronto.

43.

Anand

, Boswell

, Burton

, Hayhurst

, Harkus

, King

, Leidwinger

, McCulloch

, Meiklejohn

, Page

(2001) General guidelines for audiological practice with Indigenous Australians: Information to assist audiologists in the delivery of comprehensive, effective and appropriate audiological management for Indigenous clients/patients. Audiological Society of Australia, Forest Hill, Vic.

44.

Vos

, Barker

, Begg

, Stanley

, Lopez

(2009) Burden of disease and injury in Aboriginal and Torres Strait Islander Peoples: The Indigenous health gap. Int J Epidemiol 38, 470–477.

45.

Smith

, Flicker

, Lautenschlager

, Almeida

, Atkinson

, Dwyer

, LoGiudice

(2008) High prevalence of dementia and cognitive impairment in Indigenous Australians. Neurology 71, 1470–1473.

46.

Lo Giudice

, Smith

, Fenner

, Hyde

, Atkinson

, Skeaf

, Malay

, Flicker

(2016) Incidence and predictors of cognitive impairment and dementia in Aboriginal Australians: A follow-up study of 5 years. Alzheimers Dement 12, 252–261.

47.

Sundar

, Yu

, Sieh

, Steinbart

, Garruto

, Oyanagi

, Craig

, Bird

, Wijsman

, Galasko

, Schellenberg

(2007) Two sites in the MAPT region confer genetic risk for Guam ALS/PDC and dementia. Hum Mol Genet 16, 295–306.

48.

Battiste

, Youngblood

(2000) Protecting Indigenous knowledge and heritage: A global challenge, UBC Press.

49.

Galik

, Holmes

, Resnick

(2018) Differences between moderate to severely cognitively impaired fallers versus nonfallers in nursing homes. Am J Alzheimers Dis Other Demen 33, 247–252.

50.

MacNeil-Vroomen

, Nagurney

, Allore

(2020) Comorbid conditions and emergency department treat and release utilization in multimorbid persons with cognitive impairment. Am J Emerg Med 38, 127–131.

51.

Slaughter

, Eliasziw

, Morgan

, Drummond

(2011) Incidence and predictors of excess disability in walking among nursing home residents with middle-stage dementia: A prospective cohort study. Int Psychogeriatr 23, 54–64.

52.

Thom

, Haan

, Van Den Eeden

(1997) Medically recognized urinary incontinence and risks of hospitalization, nursing home admission and mortality. Age Ageing 26, 367–374.

53.

Weller

(2000) The relation between hip fracture and Alzheimer’s disease in the Canadian national population health survey health institutions data, 1994–1995: A cross-sectional study. Ann Epidemiol 10, 461.