Black Older Adults in the Age of Biomarkers, Physical Functioning, and Genomics

Abstract

There are persistent disparities in all-cause mortality between Blacks and Whites in the United States. Black Americans also carry the greatest burden of morbidity from different diseases of aging including heart disease, stroke, hypertension, type 2 diabetes, and certain types of cancer. Health disparities research, and particularly race/ethnic comparison studies of physical health and aging, have consistently positioned Black health in frameworks of disadvantage, suggesting that regardless of the outcome, Black people are in worse states of health and well-being relative to Whites. Yet, extensive evidence suggests that there is significant within-group variability in the aging process among Black older adults. The use of biological, physical performance, and genomic data in survey settings offer new tools and insights to interrogate heterogeneity in Black health. This chapter examines indicators of biological, physical performance, and genetic markers of aging among a national sample of Black Americans ages 54+ years with the aim of addressing two questions about heterogeneity among Black older adults: (a) How do these measures vary by age and gender among Black older adults? (b) Which indicators predict health and mortality among Black older adults? The results indicate that biological, physical performance, and genomic measures of health, generally, have more variation than simple yes or no measures of a disease, condition, or diagnosis among Black older adults, providing counternarratives to the disadvantage frameworks that dominate characterizations of Black health and aging. However, bioethical challenges limit the utility of biomarkers, physical performance, and genomics measures for Black populations.

INTRODUCTION

Black American health in the United States is communicated from a framework of disadvantage, always placing them in comparison to the gold standard: White American health. This disadvantage narrative has been aggressively reproduced in the midst of the coronavirus pandemic since Black Americans have been shown to have higher mortality rates due to COVID-19. Generally, Black Americans have shorter life spans relative to White Americans. The gap in life expectancy between Blacks and Whites is 3.6 years but is projected to widen by 40% to more than 5 years due to the coronavirus pandemic (Andrasfay & Goldman, 2021). Black Americans also live a greater portion of their lives managing chronic conditions and disability, another factor contributing to higher COVID-19 mortality rates. While we keep finding new ways to say it, racism, in all its forms, is killing Black Americans. Research has been restating and replicating this in many ways over the last few decades (Lee et al., 2017; Mays et al., 2007; U. A. Mitchell et al., 2020; Williams & Mohammed, 2008).

Research on aging at the intersection of race and older adulthood has given specific attention to accelerated aging and weathering hypotheses as explanations for Black disadvantages in health (Epel, 2009; Geronimus et al., 2015; Shiels et al., 2011). Weathering is a cumulative stress perspective that focuses on the prolonged psychosocial or physical challenges experienced by socially disadvantaged groups over the life course. These challenges increase the risk of disease, early onset of chronic conditions, and ultimately, shortened life spans (Forde et al., 2019). Weathering and accelerated aging are disadvantage frameworks and provide a mechanistic story about how social conditions get “under the skin,” which results in biological and physiological wear and tear, ultimately speeding up the aging process for Black Americans. As a result, the aging and health literature has created a one-sided story about Black health—often negative and far from the complex image of how Black older adults define their own aging process. We find, in this chapter and in our work in aging and race/ethnic health disparities more broadly, that there is a need for counternarratives (Brown & Tucker-Seeley, 2018; Croom & Marsh, 2016; Whitfield et al., 2007) that better reflect the aging experience for Black older adults. We argue that current gerontological research involving Black older adults reproduces societal inequality in and of itself, with disadvantage narratives that maintain overly simplistic stories of Black health, especially at the end of life.

With the recent emergence of biological, physical functioning and genomic data in community-based national surveys, researchers now have access to objective, subclinical, and molecular measures of health at various stages in the life course. Biological, physical function, and genomic data in survey settings offer new tools and insights that individuals might not be able to report about their own health. This may be especially important for the measuring and reporting of Black older adults’ health since self-reports of disease states may underestimate and misrepresent health for groups who have been historically excluded from the healthcare system and often distrust and have less access to healthcare that provides diagnoses (Glei et al., 2014). Collecting biomarkers, physical performance, and genomic measures in a community-based survey setting means that differential access and systematic exclusion from the healthcare system across respondents does not bias the accuracy of data toward Whiteness or White healthcare norms, which often include regular access to healthcare and physicians. While biological, physical and genomic measures of health are well positioned to quantify accelerated aging and weathering, they also generally have more intra- and interindividual variation than simple yes or no measures of a disease, condition, or diagnosis. This variability provides a much more complex picture of Black health across age, evading the Black disadvantage narrative, and providing opportunities for understanding Black heterogeneity in the aging process. The addition of physical performance, biomeasures and genomics in population surveys are often also the earliest markers of disease, disability, and mortality, providing points of intervention and prevention. Nationally representative surveys like the Health and Retirement Study (HRS) are now collecting biological data in Black older adults’ homes, making this data available to both participants and scientists at the intersections of race and aging.

This chapter examines indicators of physiological dysregulation and physical performance among a representative national sample of Black Americans aged 54 years and older, with the aim of addressing two questions about Black health and aging: (a) How do these measures vary by age and gender? and (b) Which indicators predict mortality among Black older adults? We use biomarkers, physical functioning and genomic measures in a community-based, nationally representative sample to explore new perspectives in Black health and aging. This chapter also challenges current research methodology that does not engage bioethical theory, meet an actual need of Black participants, and perpetuates knowledge or data extraction among Black older adults.

BACKGROUND

Biomarkers

The process of biological aging refers to the physiological changes that occur with age that may increase the risk for many diseases, such as cardiovascular disease, cancer, neurodegenerative disorders, and mortality (Crimmins, 2020; Crimmins, Guyer, et al., 2008; Crimmins, Vasunilashorn, et al., 2008). Beyond mortality and disease states, biomarkers that indicate functioning in major biological systems can be used as objective indicators of disease management, organ deterioration, severity, functioning loss, fragility, and disability. One advantage of the addition of biomarkers in community-based longitudinal studies of aging among older adults is that they are measured consistently across individuals and can be monitored over time without interaction with the healthcare system (Glei et al., 2014).

This chapter focuses on biomarkers that are indicative of cardiovascular functioning, metabolic processes, systemic inflammation, and organ function. The most commonly used biomarkers of cardiovascular function include systolic and diastolic blood pressure, pulse pressure, and heart rate, all of which have been linked to cardiovascular disease and mortality risk (Protogerou et al., 2007; Tverdaletal., 2008; Weiss etal., 2009). Biomarkers of metabolic processes most often include total and high-density lipoprotein (HDL) cholesterol and glycosylated hemoglobin (HbA1c), which have been shown to be associated with cardiovascular disease and mortality (Tuikkala et al., 2010; Twito et al., 2013). HDL has also been shown to be protective against heart disease (Barter et al., 2004). General systemic inflammation, commonly measured using C-reactive protein (CRP), are associated with increased risk of cardiovascular disease and mortality (Li et al., 2017). Cystatin C is a measure of kidney function and has been shown to predict risk of cardiovascular disease, chronic kidney disease, and mortality (Bell et al., 2009; Luo et al., 2015). These measures can be used as single indicators of biological aging or summed to create an overall measure of accelerated aging or weathering (Levine & Crimmins, 2014). These biomarkers are also particularly relevant for Black health and aging since they have also been linked to stress exposure and discrimination (Geronimus et al., 2015; Lee et al., 2017; Piazza et al., 2010; Simons et al., 2020).

Physical Performance

Biological markers of health and aging are not the only subclinical measures of the aging process available in nationally representative studies. There is growing interest in age-related changes in functional abilities as well using assessments of lung function, strength, balance, and mobility. Measures of physical performance were designed to assess older adults’ ability to perform movements required in accomplishing daily activities that are essential for healthy aging and independent living in advanced old age. Measures of physical performance may be especially important for older Black adults over individual self-reports of difficulty in performing daily activities for a couple of reasons. First, individual self-reports of functional limitations are more likely to be influenced by the external or built environment in which these daily activities take place. This may systematically misrepresent functioning for Black older adults (Ailshire & Crimmins, 2013). For instance, Black older adults often live in built environments that are less walkable which affects their self-report of their ability to walk independently in their neighborhoods (Clarke et al., 2009). An additional major advantage of measured indicators of physical ability over self-reported measures is that they allow for assessment of a fuller range of functional ability from low functioning or frail older adults to high functioning, healthy older adults. Performance assessments, therefore, provide a greater degree of heterogeneity among Black older adults across the spectrum of ability, especially for those who may not have any self-reported functional limitations or disabilities (Guralnik et al., 1994; Rantanen et al., 1999; Seeman et al., 1994). Physical performance measures are predictive of mortality and institutionalization and thus can also be a sensitive tool for detecting long-term risk in healthy Black older adults (Hardy et al., 2007; Studenski, 2011).

This chapter will examine commonly used physical performance measures of functional ability including lung function, walking speed, standing balance, and grip strength. Peak flow rate is an indicator of lung function and is related to physical and cognitive functioning and mortality (Clouston et al., 2013; Singh-Manoux et al., 2011). Walking speed is a marker of lower body mobility and older adults with slow walking speed are more likely to report lower self-rated health (Cesari et al., 2008) and are at higher risk for the onset and progression of disability, cognitive impairment, institutionalization, falls, and mortality (Abellan Van Kan et al., 2009; Cooper et al., 2010). Standing Balance, another indicator of lower body functioning, is associated with increased risk of falls, disability, and mortality (Nofuji et al., 2016; Onder et al., 2005). Hand-grip strength, a measure of upper body strength, is an indicator of overall strength and frailty in older adults (Chainani et al., 2016). Lower grip strength is associated with worse self-rated health (Sayer et al., 2006), higher levels of inflammation (Baylis et al., 2014), and increased risk of disability, hospitalization, and mortality (Cawthon et al., 2009; Cooper et al., 2010). In general, physical performance across these measures declines with age and, among Black older adults in particular, may be useful markers of heterogeneity in the aging process.

Genomic Measures

Genomic data like telomere length (TL), genotyped DNA, and epigenetic changes are new population-based measures in longitudinal studies of aging that have the potential to improve our understanding of the processes underlying health disparities, driven by improvements in prediction of disease risk and gradual improvements in precision medicine. These molecular and cellular modifications are recognized as fundamental to the human aging process and may increase risk for many diseases, such as cardiovascular, cancer, and neurodegenerative disorders, and even mortality (Fraga et al., 2007; Kulminski & Culminskaya, 2013). Genomic, genetic, and epigenetic changes depend on the complex interaction among hereditary, environmental, social, and/or stochastic factors, positioning genomic data as molecular biomarkers of aging that can be used to characterize inter- and intraindividual variability in the aging process for Black Americans.

For the purpose of this chapter, we focus on TL as a molecular biomarker of aging since it was measured along with our other biomarkers in our community-based sample of Black older adults. Telomeres are repeating DNA sequences that cap the ends of chromosomes and gradually shorten as a function of age and cellular division. The magnitude of this shortening has been linked to biological, genetic, and social factors of aging (Needham et al., 2015; Sanders et al., 2012; Sanders & Newman, 2013; Willeit et al., 2010). However, studies using genomic measures like TL and polygenic risk scores (PGRs)—which summarize genome-wide genotype data into a single variable that measures genetic risk to a disease or a trait (Lewis & Vassos, 2017)—have found that they are far more predictive in European ancestry groups than non-European ancestries. For example, PGRs have a prediction accuracy two to five times better in Whites than in East Asian and African American populations (Martin et al., 2019). Some studies have also found salivary telomere length (STL) does not predict indicators of health among Black older adults, presenting scientific challenges that limit the generalizability of TL research (Brown et al., 2017, 2018, 2020). Molecular-based biomarkers and social genomic measures like TL and PGRs may not be predictive in diverse population-based surveys, nor in samples of Black older adults (Martin et al., 2019). STL is an example of one biomarker for which the value may be worth questioning, including in large nationally representative studies of older adults if it does not help us understand either within race/ethnic variability or the mechanisms contributing to disparities. We examine STL as a molecular biomarker of within-group heterogeneity and whether it can predict health and mortality among older Black Americans.

METHODS

Sample

Data for this study come from the HRS, an ongoing nationally representative panel survey of adults older than age 50 in the United States. In 2006, a random one half of the sample was selected for an enhanced face-to-face (FTF) interview, which included assessments of physical function, dried blood spot collection, and anthropometric measurements. The other half of the sample received the enhanced FTF interview in 2008 and additionally included collection of saliva from which STL was assayed. We combine the sample from these two waves so that we have information on the full HRS sample in this age range. Physical and biomarker measurement was conducted by HRS interviewers in the homes of community-dwelling respondents. Respondents were not eligible for the FTF interview if they resided in a nursing home, had a proxy complete the interview, or were interviewed by telephone (Crimmins et al., 2008). This analysis includes 1,074 community-dwelling Black adults ages 54 and older with biomarker and physical performance measures and 509 Black adults with measured STL.

Measures

Biomarkers

Biomarker measurements were obtained from physical assessments and dried blood spot collection (Crimmins et al., 2008). Biomarkers of cardiovascular function, metabolic processes, inflammatory response, and organ function were included. We used definitions of risk based on clinical and research practice guidelines for biomarkers (Ailshire & Crimmins, 2013; Crimmins et al., 2008).

Cardiovascular function is measured with systolic and diastolic blood pressure, pulse pressure, and heart rate. Systolic blood pressure, diastolic blood pressure, and heart rate were measured using an automated blood pressure monitor with an inflated blood pressure cuff. Three measurements were taken, and values were averaged to create a mean score. Pulse pressure is the difference between average systolic and diastolic blood pressure. Based on clinical guidelines, high-risk values were above 140 mmHg on systolic blood pressure, above 90 mmHg on diastolic blood pressure, above or equal to 60 mmHg on pulse pressure, and heart rate of 90 beats per minute or faster.

Dried blood spots were assayed for five analytes, which are markers of metabolic function, inflammatory response, and organ function. Indicators of metabolic processes included total cholesterol, HDL cholesterol, and HbA1c. We considered individuals to be high-risk if their values were less than or equal to 40 mg/dL on HDL cholesterol and greater than or equal to 240 mg/dL on total cholesterol. We considered individuals high-risk on HbA1c if their values were greater than or equal to 6.6. Levels of general systemic inflammation are measured with CRE Those with CRP values greater than or equal to 3.0 mg/L were considered to be high risk. Cystatin C is an indicator of kidney function and individuals were considered high-risk with values greater than or equal to 1.55 mg/L.

Physical Performance

Indicators of physical performance included lung function, walking speed, balance, and grip strength. Detailed information on the protocols used to assess physical performance are available from the HRS (Crimmins et al., 2008). The physical performance measures, their means and ranges, definitions for high risk, and at-risk percentage are shown in Table 7.1. For each physical assessment, those who were unable to perform the tests, or those who did not complete the tests because either they or their interviewers thought it was unsafe or because of health reasons, were classified as poor functioning.

TABLE 7.1.

Descriptive Statistics for Physical Performance and Biomarker Measures Among African Americans 54 Years and Older, HRS 2006/2008

Variable	Mean	Min	Max	% High Risk
Biomarkers
Systolic blood pressure (mmHg)	137.0	72.3	222.0	41.2
Diastolic blood pressure (mmHg)	82.9	43.7	136.7	25.4
Heart rate (bpm)	73.6	39.0	130.0	8.9
Pulse pressure (mmHg)	54.1	18.7	105.3	30.8
HDL cholesterol (mg/dL)	55.4	20.1	116.0	20.7
Total cholesterol (mg/dL)	204.7	109.1	366.7	16.1
HbA1c (%)	6.2	3.6	14.3	22.5
CRP (mg/L)	6.2	0.0	100.0	51.7
Cystatin C (mg/L)	1.1	0.3	7.5	10.4
Physical performance
Walking speed (m/s)	0.7	0.1	2.0	16.8
Tandem stand (30 s)		0.0	1.0	35.6
Grip strength (kg)	2.5	2.5	74.5	21.1
Lung function (L/min)	353.3	60.0	800.0	28.4
Telomere length	1.4	0.38	4.27	16.6

Note: BP, blood pressure; CRP, C-reactive protein; HDL, high-density lipoprotein; HRS, Health and Retirement Study

Lung function was assessed using peak flow. Three measurements of peak expiratory flow were taken 30 seconds apart using the Mini-Wright peak flow meter. The maximum value was used as the lung function score. Those participants in the worst 25% for each gender were classified as poor lung functioning (Crimmins et al., 2008).

Walking speed was measured with a timed walk of 98.5 inches (2.5 meters) in length in the respondent’s home. Respondents were asked to complete the timed walk twice. The minimum value of the walk time (in seconds) was then divided by 2.5 to create a measure of walking speed in meters per second (m/s). Respondents could use walking aids (e.g., walking sticks, canes, walkers) to complete their timed walk. We considered respondents whose average walking speed was less than 0.6 m/s to be at high-risk, as done in other studies (Rantanen et al., 1999). Respondents who attempted but were unable to complete the timed walk, and therefore had no recorded walk time, were included in the high-risk category.

Balance was measured using the full-tandem timed balance test. Respondents were first asked to hold a semi-tandem stance, which is a midlevel standing balance test, in which they stood with the side of the heel of one foot touching the side of the big toe of the other foot. Respondents who could hold this position for 10 seconds were then asked to complete a full-tandem balance test. The tandem stance is similar to the semi-tandem stance except that respondents were asked to stand with the heel of one foot in front of and touching the toes of the other foot for 30 seconds. For the purpose of this analysis, those who were able to hold this stand for the full 30 seconds were considered to have completed the tandem balance test. We considered inability to perform the semi-tandem stance as high risk. Those who attempted but were unable to complete the semi-tandem balance test were also considered to be high risk on balance.

Grip strength was measured using a Smedley spring-type hand dynamometer with the respondent standing and holding the dynamometer at a 90° angle. Measures range from 0 kg to 100 kg. Two measurements were taken for each hand, alternating between the left and right hand. The maximum grip value from either hand was used in the analysis. Grip strength is substantially higher in men than women (Syddall et al., 2003), and those in the lowest 25% of grip strength have worse outcomes (Cawthon et al., 2009). Therefore, we consider men and women to be high risk if they are in the lowest 25% of strength relative to other men and women, respectively. We also considered respondents to be high risk if they attempted but were unable to complete the grip strength assessment.

Salivary Telomere Length

TL was measured from saliva (STL). Saliva was collected using an Oragene Collection Kit and all samples were stored in their original plates at −80°C. STL assays were performed by Telome Health (Cawthon, 2002) using quantitative polymerase chain reaction (qPCR), a well-validated and now widely accepted technique to measure TL, by comparing telomere sequence copy number in each respondent’s sample (T) to a single-copy gene copy number (S), resulting in a T/S ratio (Aviv et al., 2006; Cawthon, 2009). DNA samples were assayed in 96 well plates. The HRS took effort to minimize experimental variability by testing coefficient of variation (CV) for each sample based on three runs (three pairs of T and S runs) for plates 2 to 9, 11, and 13 and based on two runs for plates 1, 10, 13 to 64. Samples that had smaller than 12.5% CV were considered as pass and samples with greater than 12,5% CV were re-assayed (overall pass rate >98%). The mean human DNA concentration for the sample is 65.12 meg (SD = 71.5; n = 6,092). The HRS provides plate numbers in order to account for this variation in plate assay and dilution methods. We considered respondents to be high risk if they were in the shortest quartile or a T/S ratio in the lowest 25%.

RESULTS

Biomarker means, ranges, definitions for high-risk, and at-risk percentage are shown in Table 7.1.

Table 7.1 shows 52% percent of Black older adults measured high-risk on CRP, making it the most prevalent high-risk biomarker, physical performance, or genomic measure among Black older adults. The second most prevalent high-risk indictor of health was systolic blood pressure with 41% of Black older adults being high risk, followed by balance with 36% of Black older adults being considered high-risk. About a quarter of Black older adults were considered high-risk on diastolic blood pressure, pulse pressure, HDL cholesterol, HbA1c, grip strength, and lung function. Only 9% of Black older adults were measured as high risk on heart rate, making it the least prevalent high-risk indicator of health.

High-Risk By Age

We examine the age distribution of each measure in Figure 7.1. The figure plots the proportion of Black older adults who are considered high risk by age for the total sample and men and women separately. Markedly, a higher proportion of older Black Americans are at high risk at older ages for pulse pressure, cystatin C, walking speed, grip strength, lung function, and STL for both men and women. Generally, there is a greater proportion of older Black men who are at high risk within each age group for these indicators compared to Black women with the exception of walking speed. All of these measures are generally considered to be indicators of physical functioning, ability, frailty, or organ function.

FIGURE 7.1.

Percent high risk on biomarker, physical performance, and salivary telomere length by age and gender: A. Cardiovascular function, B. metabolic and inflammatory function, C. physical performance, D. telomere length.

Sociodemographic Predictors of High-Risk

Table 7.2 shows the odds ratios of high risk for each biomarker and physical performance measure across age, gender, and education for older Black Americans. The odds of being high risk for each measure by age is similar to the age differences observed in Figure 7.1. Older Black women are more likely to be high risk on CRP and walking speed and less likely to be high risk on systolic blood pressure, HDL cholesterol, and STL than Black men. Black Americans who have less education have higher odds of being high risk on pulse pressure, CRP, cystatin C, walking speed, grip strength, balance, and lung function suggesting education is a more consistent predictor of high-risk functioning among Black older adults than both gender or age.

TABLE 7.2.

Sociodemographic Characteristics Associated With High-Risk Among African Americans 54 Years and Older, HRS2006/2008

	Systolic Blood Pressure	Diastolic Blood Pressure	Heart Rate	Pulse Pressure	HbA1c	HDL Cholesterol	Total Cholesterol
Age (ref. = under 65)
65–74 years	1.03	0.53**	0.49*	1.37+	1.04	0.79	0.62*
	(0.70–1.50)	(0.37–0.76)	(0.28–0.87)	(0.95–1.98)	(0.71–1.52)	(0.54–1.15)	(0.42–0.91)
75–84 years	1.11	0.35**	0.51	2.40**	1.04	0.78	0.54*
	(0.74–1.68)	(0.19–0.63)	(0.21–1.21)	(1.57–3.67)	(0.56–1.92)	(0.45–1.34)	(0.30–0.97)
85 years and older	2.65*	0.24**	0.07*	4.67**	0.61	0.71	0.52
	(1.17–6.04)	(0.08–0.68)	(0.01–0.64)	(2.18–10.03)	(0.28–1.30)	(0.29–1.74)	(0.23–1.16)
Women	0.67*	0.83	0.91	0.73	1.05	0.47**	1.12
	(0.46–0.99)	(0.56–1.22)	(0.53–1.57)	(0.50–1.09)	(0.70–1.55)	(0.27–0.83)	(0.70–1.79)
Education, years	0.97	0.99	0.91+	0.96+	1.00	0.97	0.97
	(0.93–1.02)	(0.92–1.05)	(0.82–1.00)	(0.93–1.00)	(0.94–1.07)	(0.92–1.03)	(0.90–1.04)
Observations	1,074	1,074	1,074	1,074	1,074	1,074	1,074
Age (ref = under 65)
65–74 years	1.08	1.35		1.51*	2.29**	2.51**	1.35
	(0.76–1.52)	(0.71–2.54)		(1.09–2.10)	(1.61–3.25)	(1.62–3.88)	(0.68–2.68)
75–84 years	1.17	3.82**	1.90**	3.93**	6.78**	4.61**	2.70**
	(0.75–1.80)	(2.10–6.95)	(1.23–2.94)	(2.41–6.41)	(4.41–10.42)	(2.66–7.99)	(1.38–5.28)
85 years and older	0.71	5.07**	7.55**	9.32**	9.47**	16.68**	2.66*
	(0.39–1.30)	(2.09–12.28)	(3.09–18.47)	(4.54–19.11)	(4.90–18.30)	(6.83–40.72)	(1.10–6.41)
Women	1.71**	1.36	1.43+	0.80	0.83	0.99	0.47**
	(1.18–2.47)	(0.71–2.60)	(0.98–2.09)	(0.54–1.18)	(0.57–1.21)	(0.65–1.52)	(0.28–0.79)
Education, years	0.96+	0.95+	0.93*	0.91**	0.91**	0.87**	0.92+
	(0.92–1.00)	(0.90–1.00)	(0.87–0.99)	(0.87–0.95)	(0.87–0.95)	(0.83–0.91)	(0.84–1.01)
Observations	1,074	1,074	637	1,074	1,074	1,074	509

Note: CRP, C-reactive protein; HDL, high-density lipoprotein; HRS, Health and Retirement Study.

95% Confidence Interval in parentheses.

^*p < .05.

^**p< .01.

⁺p < .1.

Predicting Mortality

Since our biomarkers of aging, physical performance, and STL measures may be predictive of health and longevity among older Blacks, we examine the odds of subsequent 10-year mortality associated with being high risk on each indicator. Because measures of blood pressure are highly correlated, pulse pressure is the only measure of blood pressure included in the models (Ailshire & Crimmins, 2013).

Table 7.3 shows the relationship between high-risk indicators of health and death among older Black Americans. These findings suggest that being at high-risk on physical performance measures, HbA1c, and Cystatin C are predictive of 10-year mortality. Interestingly, high-risk STL is marginally associated with lower odds of 10-year mortality, further complicating the utility of STL and molecular biomarkers in understanding health and aging for older Black adults (Brown etal., 2017, 2018, 2020).

TABLE 7.3.

Predictors of Mortality Among African Americans Ages 54 and Older, HRS 2006/2008

	Sociodemographic	Cardiovascular	Metabolic, Inflammatory	Performance	Telomere Length	All High-Risk Indicators
Age (ref. = under 65)
65–74 years	1.56*	1.56*	1.50*	1.13	1.68+	1.06
	(1.10–2.21)	(1.09–2.23)	(1.06–2.13)	(0.77–1.66)	(0.96–2.95)	(0.71–1.59)
75–84 years	3.19**	3.09**	2.65**	1.59*	4.10**	1.31
	(2.17–4.68)	(2.06–4.62)	(1.77–3.96)	(1.01–2.52)	(2.34–7.19)	(0.80–2.14)
8.5 years and older	7.07**	6.50**	5.59**	2.53**	7.25**	1.88*
	(4.58–10.93)	(4.01–10.53)	(3.52–8.87)	(1.44–4.44)	(3.49–15.06)	(1.01–3.50)
Women	0.67**	0.69*	0.63**	0.69*	0.64*	0.65**
	(0.51–0.89)	(0.52–0.92)	(0.47–0.84)	(0.52–0.93)	(0.42–0.98)	(0.48–0.87)
Education, years	0.95**	0.95*	0.95*	0.98	0.92*	0.98
	(0.91–0.98)	(0.91–0.99)	(0.91–0.99)	(0.94–1.02)	(0.85–0.99)	(0.94–1.03)
High-risk
Heart rate		1.43				1.31
		(0.88–2.33)				(0.91–1.87)
Pulse pressure		1.35+				1.33
		(0.99–1.83)				(0.93–1.89)
HbA1c			1.50*			1.76**
			(1.10–2.06)			(1.21–2.55)
HDL cholesterol			0.72			2.12**
			(0.49–1.07)			(1.43–3.13)
Total cholesterol			0.94			1.25
			(0.64–1.40)			(0.77–2.04)
CRP			0.99			1.29
			(0.73–1.36)			(0.95–1.73)
Cystatin C			2.69**			1.43*
			(1.88–3.85)			(1.05–1.96)
Walking speed				1.24		0.70+
				(0.87–1.77)		(0.48–1.03)
Grip strength				1.31		0.92
				(0.91–1.89)		(0.61–1.39)
Balance				1.86**		0.95
				(1.26–2.73)		(0.69–1.30)
Lung function				2.18**		2.37**
				1.24		(1.66–3.37)
Telomere length					0.67
					(0.41–1.08)
Observations	1,074	1,074	1,074	1,074	509	1,074

Note: CRP, C-reactive protein; HDL, high-density lipoprotein; HRS, Health and Retirement Study.

95% Confidence Interval in parentheses.

^*p < .05.

^**p < .01.

⁺p < .1.

DISCUSSION

One major goal of this chapter was to support the broadening of approaches and measures used in the study of Black older adults by suggesting that Black/White comparisons alone are not the answer to understanding Black health and aging. The heterogeneity in health and functioning among Black older adults challenges the disadvantage narrative that often characterizes Black health. If we had only focused our analyses on Black/White differences, using Whites as the gold standard of health, we would miss out on the variability that exists within our sample of Black older adults. There are good reasons for using multiple measures and approaches in conceptualizing Black health and aging that go beyond self-reports and between group comparisons (Taylor & Chatters, 2020). Large community-based, nationally representative samples of older adults provide an ideal setting to collect physical performance, biomarker, and genomic data for understanding the heterogeneity in health and longevity among Black populations (Mitchell et al., 2018).

Our findings among older Black adults indicate physical functioning measures show the strongest declines with age including walking speed, grip strength, and lung function. Only a couple of the biomarkers in our sample decline with age including pulse pressure and cystatin C. STL also declined with age after age 80. Yet, indicators of metabolic and cardiovascular function are not consistently related to age in this sample of Black older adults, with the exception of pulse pressure. Other research has shown that pulse pressure may be the best indicator of the risk posed by high blood pressure (Ailshire & Crimmins, 2013; Weiss et al., 2009). Future work should consider the ability of metabolic and cardiovascular biomarkers in predicting diagnosis of cardiovascular disease and medication use among Black older adults. Gender and education are both correlates of biomarker, physical performance, and STL for older Black adults but education is a more consistent predictor of high-risk functioning than gender or age.

We also show the utility of biomarkers, physical performance, and genomic measures of aging in predicting 10-year mortality among Black older adults. These findings suggest that physical performance measures—including walking speed, balance, and lung function—are better predictors of 10-year mortality than metabolic, inflammatory, or organ specific biomarkers. The only biomarkers that predict mortality are HbA1c (a marker of glucose levels and insulin intolerance) and Cystatin C (a marker of kidney function). Thus, studies aiming to incorporate objective measures of health and aging among older Black adults may want to prioritize the collection of physical performance measures since they are, overall, predictors of mortality in this sample of Black older adults as well as in older populations more generally (Ailshire & Crimmins, 2013). Given that there are few biomarkers that predict mortality among older Black adults, it is important that researchers consider the appropriate biomarker that assesses the particular health condition, outcome, or process that they intend (Best & Chenault, 2014).

Our singular molecular biomarker, STL, is marginally associated with lower odds of 10-year mortality, further complicating the utility of STL and molecular biomarkers in understanding health and aging for older Black adults (Brown et al., 2017, 2018, 2020). Similar to other social genomics measures, this counterintuitive finding suggests STL may not be a useful clinical marker of functional aging among Black older adults. TL is a highly heritable trait, with estimates ranging from 0.36 to 0.82 (Andrew et al., 2006; Honig et al., 2015), denoting the proportion of the phenotypic population variation that is due to genotypic variation. Thus, future analyses may consider incorporating genetic ancestry since TL is highly heritable and it has been shown to improve predictive capacity of disease by accounting for within-group heterogeneity that is not captured by social categorizations of race/ethnicity (Galanter et al., 2017; Kumar et al., 2010; Nalls et al., 2008; Udler et al., 2015). Additionally, individual rate of change in TL may be more important in capturing the variability in health and functioning among older Black adults than absolute TL since Black people may have longer TL over the life course but experience faster shortening (Aviv et al., 2006; Brown et al., 2017; Hunt et al., 2015). Longitudinal data for biomarkers, physical performance, and genomic measures are needed and ideal to fully appreciate differences with age; this may be especially true for molecular markers like STL.

Community Engagement and Bioethics

Improving the translational utility of biomarker, physical performance, and genomic data for Black communities depends largely on the inclusion of Black populations in research. Inclusion, however, can only come with the cooperation of individuals and communities in granting access for the sharing and use of their data and samples. Engaging bioethical theory in the collection of biomarker, physical performance, and genomic data practices among older Black populations provides an opportunity to promote individual and community autonomy in the governing, collection, ownership, and application of data. Community-based data collection methods provide an opportunity to not just extract data, but to directly connect Black older adults to needed healthcare services and resources, subsidized medications, royalties, or intellectual-property rights (Fox, 2020). Ethical, transparent, and symbiotic data practices and resources sharing are an important step in restoring and obtaining trust and participation in research among Black communities. Data collection frameworks need to ensure that data and biospecimens can be shared without harm to participants and that there is an expectation of and mutually negotiated plans for community and individual benefit sharing. In addition, fully engaging with the diversity within Black communities in biomarker, physical performance, and genomics research also rejects simplistic scenarios that reduce complex social, economic, and political histories of oppression and discrimination into deterministic race scholarship (Mitchell, Perry, Johnson-Lawrence, et al., 2020; Mitchell, Perry, Rorai, et al., 2020; Pereira et al., 2021). Due to the nature and history of scientific racism in biological and genetic research, special attention and acknowledgment of this history, combined with community engagement and research partnerships with Black older adults will help improve the quality and utility of these data.

Measurement Considerations: Differential Consent, Missingness, and Selective Mortality

Differential consent, missingness, and selective mortality can influence results from biomarker, physical performance, and genomic analyses and that is especially true for Black older adults. Not all respondents complete the physical performance, biomarker and genomic measurements. The analytic sample did not include respondents who did not give consent to participate in the physical assessment, blood draw, or saliva sample and also excluded those who consented but were unable to complete the tests or for whom blood assays or physical performance measures could not be obtained. Black older adults with no measured values of health and functioning due to differential consent or missingness may differ in important ways from the Black older adults who are in the analytic sample. Respondents with missing or incomplete blood draw data often have lower cognitive function and more disability (Ailshire & Crimmins, 2013). Selective mortality—or the often unequal distribution of respondents who survive long enough to the time of data collection—can also lead to bias, especially for Black populations who have higher mortality rates from most leading causes of death at younger ages (Zajacova & Burgard, 2013).

Although they offer major advantages over self-reports, there are some additional limitations to using measured markers of health and functioning in Black populations. First, individuals who are incapable of participating in performance assessments because of physical or cognitive impairment are not included. Thus, performance assessments may do a better job differentiating ability among Black older adults with higher levels of function. Physical ability also cannot be assessed among Black older adults who have health conditions (e.g., surgery or injury) that prevent them from being able to participate in physical performance tests. Second, Black older adults may be more likely to refuse consenting to a blood draw or participating in a physical performance assessment than they are to refuse answering questions about their physical limitations and health conditions. Finally, the successful collection of measured markers of health and functioning is dependent on being in an appropriate location (e.g., space for a timed walk or location for a blood draw) and having equipment that functions correctly (e.g., blood pressure cuff). Using the combination of reported and measured health and functioning may be the best way to characterize the heterogeneity in Black health and aging.

CONCLUSION

The inclusion of biomarkers, physical performance, and social genomics measures in population research has the potential to improve our understanding of the aging process among Black Americans. When White older adults are not considered the gold standard of health to whom we compare Black older adults, we find there is heterogeneity in the aging process among these older adults. The variability in health and aging found among Black older adults in this chapter serves as a counternarrative to disadvantage frameworks about Black health. Physicians and researchers aiming to identify Black older adults at increased risk of mortality in the next 10 years should consider using validated physical performance measures like walking speed, balance, and lung function. These are all simple, cheap, and have established predictive capacity, especially in identifying frailty among older adults and especially among Black older adults. Biomarkers, physical performance, and genomics measures in studies of aging may play an important role in advancing the basic understanding of pathways and mechanisms to disease states, disability, and death among Black older adults. Further research should investigate the utility of these markers in identifying Black older adults in need of intervention and connection to health services. Finally, researchers must be implicated in cocreating ways of ethically engaging Black older adults in research that includes individual and community level benefits sharing.