Healthy Aging in the Context of Educational Disadvantage: The Role of “Ordinary Magic”

Abstract

Objective: The objective of this study is to examine the correlates of healthy aging in the context of educational disadvantage and the extent to which identified correlates are shared with the wider, more educationally advantaged population. Method: Data are from the 2012 Health and Retirement Study. The analytic sample included 17,484 self-respondents ≥50 years of age. Educational disadvantage was defined as having less than a high school diploma. Using logistic regression, healthy aging was regressed on demographic, early-life, and health-related factors by educational status. Results: Among educationally disadvantaged adults, demographic characteristics (e.g., age), health practices (e.g., physical activity), and the presence of health conditions were independently correlated with healthy aging. With few exceptions, correlates of healthy aging were similar among educationally advantaged and disadvantaged adults. Discussion: Ordinary factors are associated with healthy aging among adults without a high school diploma, suggesting that healthy aging is possible for larger numbers of adults aging in the context of educational disadvantage.

Keywords

health behavior resilience socioeconomic status successful aging

Introduction

Education is a key determinant of health. Individuals with higher levels of education report better self-rated health (Liu & Hummer, 2008), are less likely to develop late-life disability (Taylor, 2010), and typically live longer (Meara, Richards, & Cutler, 2008) than their counterparts with lower levels of education. Education is believed to influence a host of factors that impact health (e.g., knowledge, employment opportunities, working conditions, financial resources, health practices, and exposure to stressors; Braveman, Egerter, & Williams, 2011), with adults with less education less favorably positioned to maintain their health than those with higher education.

One health outcome for which educational differences are evident is healthy or “successful” aging. In their examination of successful aging among participants of the Alameda County Study, Strawbridge, Wallhagen, and Cohen (2002) reported that 10.8% of adults with less than 12 years of schooling were aging “successfully” compared with 17.2% of high school graduates and 21.7% of those with more than 12 years of schooling. McLaughlin, Connell, Heeringa, Li, and Roberts (2010) reported a similar pattern by education among older U.S. adults in each of four time points between 1998 and 2004.

Notwithstanding the fact that healthy or successful aging occurs less frequently among those with less education, some educationally disadvantaged adults manage to maintain robust health into older adulthood (e.g., 10.8% of those with <12 years of schooling in Strawbridge et al.’s aforementioned study). Little is known, however, about the sociodemographic characteristics of those who manage to do so or the specific factors that may be responsible for their robust health. Existing work on resilience offers insight.

Resilience has been defined as “a class of phenomena characterized by good outcomes in spite of serious threats to adaptation or development” (Masten, 2001, p. 228, emphasis in original). In scholarly discussion of resilience, Masten (2001) reported that it was once thought that children who flourished in the face of adversity had “special qualities” that rendered them immune to the hazards to which they were exposed. She explained, however, that empirical research on resilience suggests something to the contrary—that “resilience does not come from rare and special qualities, but from the everyday magic of ordinary, normative human resources in the minds, brains, and bodies of children, in their families and relationships, and in their communities” (Masten, 2001, p. 235). With respect to healthy aging, this suggests that the factors that promote healthy aging among educationally disadvantaged adults may not be a great mystery. Instead, “ordinary” factors (e.g., health practices) are likely to explain healthy aging in the context of educational disadvantage.

Considerable effort has been directed at understanding inequalities in late-life health by socioeconomic factors such as education, with the existing body of research dominated by between-group comparisons of advantaged and disadvantaged individuals. As emphasized more than a decade ago by the Committee on Future Directions for Behavioral and Social Sciences Research at the National Institutes of Health (2001), however, efforts are also “needed to identify protective factors among low SES individuals who defy the odds of succumbing to ill health” (p. 49). Such efforts have the potential to reduce socioeconomic inequalities in health by advancing knowledge of what works within subgroups at elevated risk of poor health outcomes.

The objective of this investigation is to advance understanding of healthy aging in the context of educational disadvantage by examining potential correlates of healthy aging among educationally disadvantaged adults and the degree to which identified correlates are similar to those associated with healthy aging in the wider, more educationally advantaged population. The latter will allow us to determine if certain factors may exert a differential effect in the context of educational disadvantage.

Method

Data Source and Sample

Data for this investigation are from the 2012 wave of the Health and Retirement Study (HRS), a national study of U.S. adults age 50 and over (Juster & Suzman, 1995). In 2012, 20,554 adults completed an in-person or telephone interview. Of those, 3,070 were excluded from the analysis because they had sampling weights of zero (n = 1,703), required a proxy respondent (n = 810), were of “other” race/ethnicity (n = 556), or were missing imputed cognitive data (n = 1). The resulting analytic sample included 17,484 individuals.

Measures

Educational status

HRS participants were divided into two educational strata: those without a high school diploma and those with a high school or higher degree. Educational disadvantage was defined as having earned less than a high school diploma.

Healthy aging

Informed by McLaughlin, Jette, and Connell (2012), three criteria were used to determine healthy aging status: absence of cognitive impairment, freedom from disability, and high physical functioning. Those meeting all three criteria were classified as experiencing healthy functional aging.

Absence of cognitive impairment

A cognitive score was calculated for each respondent by summing scores on four cognitive tasks: an immediate word recall, delayed word recall, serial subtraction test, and backward count. Respondents obtaining a score of 12 or higher (out of 27) were classified as being free of cognitive impairment (for additional details on the measures and cut-point, see Crimmins, Kim, Langa, & Weir, 2011).

Freedom from disability

Individuals reporting no difficulty performing six activities of daily living (e.g., dressing) and five instrumental activities of daily living (e.g., managing money) because of a health or memory problem were classified as being free of disability (see Table 2 for a complete list of activities).

High physical functioning

Respondents were asked if they have difficulty performing 11 physical tasks (e.g., walking one block, climbing one flight of stairs) because of a health problem. Individuals reporting difficulty with no more than one task were defined as having high physical functioning (see Table 2 for a complete list of tasks).

Correlates of healthy aging

Guided by existing literature, a range of potential correlates of healthy functional aging were examined, including demographic factors, early-life characteristics, health behaviors, and physical and mental health conditions.

Demographic factors

Age category (middle-aged: 50 to 64, young-old: 65 to 74, old-old: 75 to 84, oldest-old: 85+), gender, marital status (married or living with a partner vs. not married or partnered; hereinafter referred to as married or not married), race/ethnicity (Black, Hispanic, White), and quartiles of total household wealth (calculated separately for each educational stratum) were examined. In the HRS, a household consists of the respondent and—if married or partnered—his or her spouse/partner (HRS, 2008); total household wealth refers to the sum of the household’s assets minus debt, including housing (for additional details on wealth components, see Chien et al., 2015).

Early-life characteristics

Self-rated childhood health was ascertained based on responses to the following item: “Consider your health while you were growing up, before you were 16 years old. Would you say that your health during that time was excellent, very good, good, fair, or poor?” Response options were collapsed, creating a dichotomous indicator of childhood health (i.e., fair or poor vs. good, very good, and excellent). Early-life financial status was also assessed. Participants were asked, “Now think about your family when you were growing up, from birth to age 16. Would you say your family during that time was pretty well off financially, about average, or poor?” Childhood families were categorized as poor or not poor (i.e., pretty well off and about average).

Health behaviors

Respondents were asked if they “ever drink any alcoholic beverages such as beer, wine, or liquor?” Those answering “yes” were classified as current users of alcohol. Using self-reported height and weight, body mass index (BMI) was categorized as normal/underweight (<25; n = 248 for underweight), overweight (≥25, <30), or obese (≥30). Physical activity was assessed with two items: “How often do you take part in sports or activities that are vigorous, such as running or jogging, swimming, cycling, aerobics or gym workout, tennis, or digging with a spade or shovel?” and “How often do you take part in sports or activities that are moderately energetic such as, gardening, cleaning the car, walking at a moderate pace, dancing, floor or stretching exercises.” Response options for both items included more than once a week, once a week, one to three times a month, and hardly ever or never. Individuals engaging in either moderate or vigorous physical activity more than once a week were classified as physically active. Respondents who indicated that they ever smoked were classified as ever smokers; all others were classified as never smokers.

Physical and mental health conditions

A variable capturing total number of chronic conditions was created from seven items asking participants if a doctor had ever told them that they have arthritis, cancer, chronic lung disease, diabetes, heart disease, hypertension, or stroke. Depressive symptoms were assessed with an eight-item Center for Epidemiologic Studies Depression (CES-D) scale. Respondents reporting four or more symptoms were classified as having a clinically significant level of depressive symptoms (Steffick, 2000).

Data Analysis

Following calculation of descriptive statistics, a series of multiple logistic regression models were used to examine the association between each potential correlate and healthy aging by educational status. In Model 1, healthy aging was regressed on age, gender, marital status, race and ethnicity, and total household wealth. Early-life characteristics were added in Model 2, followed by the addition of health behaviors in Model 3. Physical and mental health conditions were introduced in Model 4.

Values for missing cognitive and wealth data were imputed prior to release of the data sets used in this investigation. IVEware (Raghunathan, Solenberger, & Van Hoewyk, 2002) was used to multiply impute missing values for the remaining variables; no greater than 5% of data were imputed for any variable.

Using the respondent level weights provided by HRS, data were weighted to reflect the U.S. population of middle-aged and older adults in 2012. Standard errors were adjusted for clustering and stratification, two features of the sampling design used to select HRS participants. All analyses were completed using SAS Version 9.3 and IVEware Version 0.1.

Results

Sample Characteristics

Of the individuals included in this investigation, 3,203 (18.3%) had less than a high school diploma and 14,281 (81.7%) had a high school or higher degree. As evident in Table 1, the age distribution was older for adults without a high school diploma than for those with a high school or higher degree (61.9% vs. 44.4% were 65+). The percentage of women was similar across educational strata; however, there were notable differences in marital status, race/ethnicity, and total household wealth. Among those without a high school diploma, just over half (53.0%) were married, 51.1% were members of a racial/ethnic minority group, and total household wealth averaged less than $145,000. In contrast, two thirds of those with a high school or higher degree were married, nearly 86% were non-Hispanic White, and total household wealth averaged more than $550,000. Early-life characteristics also varied across educational strata. Compared with their counterparts with a high school or higher degree, a greater percentage of adults with less than a high school diploma reported poor or fair childhood health (11.3% vs. 4.9%) and that their childhood family was poor (46.3% vs. 22.8%).

Table 1.

Sample Characteristics by Level of Education.

Characteristics	Less than high school diploma (n = 3,203)^a	High school diploma or higher education (n = 14,281)^a
Characteristics	M or % [95% CI]	M or % [95% CI]
Demographic
Age category (years)
50 to 64	38.2 [33.5, 42.9]	55.6 [52.9, 58.3]
65 to 74	30.9 [27.7, 34.1]	25.9 [24.3, 27.5]
75 to 84	21.3 [18.9, 23.6]	13.7 [12.6, 14.8]
85 and over	9.7 [8.0, 11.3]	4.8 [4.2, 5.4]
Female	56.7 [54.8, 58.6]	54.5 [53.7, 55.3]
Married	53.0 [50.8, 55.3]	66.3 [64.9, 67.7]
Race/ethnicity
Hispanic	32.5 [25.3, 39.6]	5.3 [4.2, 6.4]
Non-Hispanic Black	18.6 [15.4, 21.8]	9.1 [8.0, 10.3]
Non-Hispanic White	48.9 [42.7, 55.1]	85.6 [84.0, 87.1]
Total household wealth ($), mean	141,432 [114,380, 168,484]	552,089 [506,484, 597,695]
Early-life
Fair/poor childhood health	11.3 [9.9, 12.7]	4.9 [4.4, 5.4]
Poor childhood family	46.3 [43.7, 48.9]	22.8 [21.9, 23.6]
Health
Alcohol use	39.7 [37.7, 41.7]	62.9 [61.3, 64.6]
Body mass index
Normal/underweight	26.8 [24.6, 28.9]	28.7 [27.7, 29.7]
Overweight	37.6 [34.7, 40.6]	36.9 [35.9, 38.0]
Obese	35.6 [33.0, 38.2]	34.4 [33.2, 35.5]
Ever smoker	61.8 [59.0, 64.5]	55.4 [54.1, 56.7]
Physically active	41.5 [38.8, 44.3]	60.4 [59.0, 61.8]
CES-D score ≥ 4	25.8 [23.1, 28.5]	12.2 [11.5, 12.9]
Number of chronic health conditions, mean	2.3 [2.2, 2.4]	1.8 [1.8, 1.9]^b

Note. Means and percentages are weighted. Sample numbers are not weighted. Percentages may sum to more than 100 owing to rounding. CI = confidence interval. CES-D = Center for Epidemiologic Studies Depression.

Average n across imputed data sets.

Confidence interval includes point estimate owing to rounding.

The percentage of adults reporting alcohol use (39.7% vs. 62.9%) and regular physical activity (41.5% vs. 60.4%) was notably lower for those with less than a high school diploma than for those with a high school or higher degree. Differences were also evident for smoking history, with 61.8% of those without a diploma reporting ever smoking compared with 55.4% of those with a high school or higher degree. Educational differences in BMI were minimal, with over 70% of adults in both educational strata having a BMI indicative of overweight or obesity. Average number of chronic health conditions (2.3 vs. 1.8) and the percentage reporting four or more depressive symptoms (25.8% vs. 12.2%) were higher for those without a high school diploma.

Healthy Aging

The percentage of adults experiencing healthy aging was significantly lower among those with less than a high school diploma (18.6%) than for those with a high school or higher degree (48.9%; p < .0001). As displayed in Table 2, this same general pattern was evident for each subcomponent of healthy aging.

Table 2.

Prevalence of Healthy Aging and Its Components by Level of Education.

	Less than high school diploma	High school diploma or higher education
	% [95% CI]	% [95% CI]
Healthy aging	18.6 [16.8, 20.5]	48.9 [47.2, 50.6]
Healthy aging components
Absence of cognitive impairment	50.5 [48.1, 52.9]	87.8 [87.1, 88.5]
No disability^a	63.9 [61.8, 66.0]	82.7 [81.8, 83.6]
High physical functioning^b	32.1 [30.3, 33.8]	54.2 [52.5, 55.8]

Note. Percentages are weighted. CI = confidence interval.

No self-reported difficulty with the following tasks because of health or memory problem: walking across a room; dressing, including putting on shoes and socks; bathing or showering; eating, such as cutting up food; getting in or out of bed; using the toilet, including getting up and down; preparing a hot meal; shopping for groceries; making phone calls; taking medications; managing money, such as paying bills and keeping track of expenses.

No greater than one self-reported difficulty with the following tasks because of a health problem: walking one block; walking several blocks; sitting for about 2 hr; getting up from a chair after sitting for long periods; climbing one flight of stairs without resting; climbing several flights of stairs without resting; stooping, kneeling, or crouching; reaching or extending arms above shoulder level; pulling or pushing large objects like a living room chair; lifting or carrying weights over 10 pounds, like a heavy bag of groceries; picking up a dime from a table.

Correlates of Healthy Aging

After simultaneously controlling for other demographic factors, age, gender, race/ethnicity, and total household wealth were significantly correlated with healthy aging in both educational strata (see Model 1 in Table 3). Compared with middle-aged adults, the odds of healthy aging were 31% to 53% lower for “young-old” adults, 68% to 78% lower for “old-old” adults, and 88% to 91% lower for “oldest-old” adults, depending on the educational stratum. The difference between middle-aged and young-old adults failed to reach statistical significance, however, in those without a high school diploma (p = .093). Relative to their male counterparts, the odds of healthy aging were reduced by roughly 30% among women in both educational strata. Differences by total household wealth were also evident. Compared with those in the fourth (i.e., highest) quartile of household wealth, the odds of healthy aging were 22% to 37% lower for those in the third quartile, 52% to 61% lower for those in the second quartile, and 66% to 78% lower for those in the first quartile, depending on the educational stratum. Racial and ethnic differences were also observed. Compared with their non-Hispanic White counterparts, the odds of healthy aging were 30% to 43% lower for Black adults, depending on the educational stratum. Patterns diverged by education for Hispanic ethnicity. Relative to non-Hispanic White adults, the odds of healthy aging were 40% higher for Hispanic adults without a diploma and 22% lower for Hispanic adults with a high school or higher degree. Last, a borderline significant association was evident for marital status among those without a diploma (p = .051), with the odds of healthy aging 37% higher for married adults relative to their unmarried counterparts.

Table 3.

Odds Ratios [95% Confidence Intervals] for Healthy Aging by Level of Education.

Variables	Less than a high school diploma				High school diploma or higher education
Variables	Model 1	Model 2	Model 3	Model 4	Model 1	Model 2	Model 3	Model 4
Demographic
Age category (Referent = middle-aged)
Young-old: 65 to 74	0.69 [0.45, 1.06]	0.70 [0.45, 1.08]	0.73 [0.46, 1.13]	1.00 [0.66, 1.51]	0.47*** [0.42, 0.52]	0.47*** [0.42, 0.53]	0.51*** [0.46, 0.57]	0.67*** [0.60, 0.76]
Old-old: 75 to 84	0.32*** [0.22, 0.46]	0.32*** [0.22, 0.47]	0.32*** [0.22, 0.47]	0.48*** [0.33, 0.70]	0.22*** [0.19, 0.25]	0.23*** [0.20, 0.26]	0.24*** [0.21, 0.28]	0.38*** [0.33, 0.45]
Oldest-old: 85 and over	0.12*** [0.06, 0.27]	0.13*** [0.06, 0.27]	0.12*** [0.06, 0.26]	0.17*** [0.08, 0.36]	0.09*** [0.07, 0.12]	0.09*** [0.07, 0.12]	0.10*** [0.07, 0.13]	0.15*** [0.11, 0.20]
Female	0.70* [0.54, 0.92]	0.71* [0.53, 0.93]	0.84 [0.63, 1.14]	1.02 [0.74, 1.40]	0.67*** [0.61, 0.74]	0.67*** [0.61, 0.74]	0.68*** [0.62, 0.75]	0.65*** [0.59, 0.72]
Married	1.37 [1.00, 1.89]	1.35 [0.99, 1.84]	1.44* [1.06, 1.95]	1.36 [1.00, 1.86]	1.05 [0.93, 1.18]	1.05 [0.94, 1.18]	1.07 [0.94, 1.20]	1.00 [0.88, 1.14]
Race/ethnicity (Referent = Non-Hispanic White)
Hispanic	1.40* [1.01, 1.93]	1.49* [1.06, 2.09]	1.20 [0.82, 1.75]	1.05 [0.71, 1.57]	0.78* [0.61, 0.99]	0.84 [0.66, 1.07]	0.86 [0.68, 1.07]	0.79* [0.63, 0.99]
Non-Hispanic Black	0.57*** [0.42, 0.78]	0.58*** [0.42, 0.79]	0.55*** [0.40, 0.76]	0.49*** [0.35, 0.69]	0.70*** [0.60, 0.82]	0.74*** [0.63, 0.86]	0.84* [0.73, 0.98]	0.82* [0.71, 0.96]
Total household wealth (Referent = fourth quartile)
First quartile	0.34*** [0.23, 0.50]	0.35*** [0.23, 0.51]	0.41*** [0.28, 0.60]	0.56** [0.37, 0.84]	0.22*** [0.19, 0.26]	0.23*** [0.20, 0.27]	0.34*** [0.29, 0.40]	0.41*** [0.35, 0.48]
Second quartile	0.48** [0.30, 0.76]	0.49** [0.31, 0.77]	0.60* [0.38, 0.95]	0.71 [0.45, 1.11]	0.39*** [0.34, 0.46]	0.40*** [0.35, 0.47]	0.54*** [0.46, 0.63]	0.61*** [0.52, 0.71]
Third quartile	0.78 [0.52, 1.17]	0.81 [0.54, 1.20]	0.96 [0.64, 1.44]	1.03 [0.68, 1.56]	0.63*** [0.54, 0.73]	0.64*** [0.54, 0.74]	0.73*** [0.63, 0.85]	0.76*** [0.65, 0.89]
Early-life
Fair or poor childhood health		0.61* [0.38, 0.97]	0.56* [0.35, 0.88]	0.66[0.39, 1.11]		0.48*** [0.38, 0.61]	0.52*** [0.41, 0.65]	0.64*** [0.50, 0.81]
Poor childhood family		0.73* [0.57, 0.94]	0.75* [0.57, 0.98]	0.90[0.69, 1.19]		0.79*** [0.71, 0.87]	0.81*** [0.73, 0.90]	0.85** [0.76, 0.95]
Health behaviors
Body mass index (Referent = normal/underweight)
Overweight			0.77[0.56, 1.06]	0.84[0.59, 1.21]			0.89[0.78, 1.01]	0.99[0.86, 1.13]
Obese			0.56** [0.39, 0.80]	0.68* [0.48, 0.98]			0.52*** [0.46, 0.59]	0.70*** [0.61, 0.80]
Ever smoker			0.74[0.53, 1.05]	0.90[0.64, 1.26]			0.69*** [0.62, 0.78]	0.77*** [0.69, 0.87]
Physically active			1.87*** [1.53, 2.27]	1.58*** [1.30, 1.92]			2.45*** [2.21, 2.72]	2.07*** [1.86, 2.31]
Uses alcohol			2.10*** [1.62, 2.72]	1.86*** [1.45, 2.39]			1.55*** [1.41, 1.70]	1.38*** [1.25, 1.52]
Health conditions
Number of chronic health conditions				0.60*** [0.54, 0.68]				0.60*** [0.57, 0.63]
CES-D score ≥ 4				0.29*** [0.17, 0.51]				0.28*** [0.24, 0.34]

Note. CES-D = Center for Epidemiologic Studies Depression.

p < .05. **p < .01. ***p < .001.

Model 2 in Table 3 includes early-life characteristics. Among adults without a diploma, the odd of healthy aging were 39% lower for those reporting fair or poor childhood health relative to those who did not; the respective figure was 52% for those with a high school or higher degree. A significant association was also evident for childhood socioeconomic status (SES).Depending on the educational stratum, the odds of healthy aging were 21% to 27% lower among adults reporting that they grew up in a poor childhood family. Patterns observed by age, gender, marital status, and total household wealth remained largely unchanged after controlling for early-life characteristics. With the exception that the odds of healthy aging were no longer significantly different for Hispanic and White adults among those with a high school or higher degree (p = .155), patterns by race and ethnicity were also similar.

Model 3 includes health behaviors. In both educational strata, alcohol use and physical activity were significantly associated with healthy aging. Relative to their respective counterparts, the odds of healthy aging were 1.6 to 2.1 times higher among adults who reported alcohol use and 1.9 to 2.5 times higher for adults who reported engaging in moderate or vigorous physical activity more than once a week, depending on the educational stratum. Elevated BMI and ever smoking were also associated with healthy aging. Depending on the educational stratum, the odds of healthy aging were 44% to 48% lower for older adults with a BMI indicative of obesity (relative to normal/underweight BMI). Ever smoking reduced the odds of healthy aging by 26% among those without a diploma and 31% among those with a high school or higher degree; however, the difference was not statistically significant among those without a diploma (p = .091). After controlling for behavioral factors, gender (p = .269) and Hispanic ethnicity (p = .342) were no longer significantly associated with healthy aging in those without a diploma. Marital status attained statistical significance, however, with the odds of healthy aging 44% higher for married adults among those without a diploma (p = .019).

Physical and mental health conditions were added in Model 4, with observed effect sizes virtually identical across educational strata. Specifically, each increase in the number of chronic conditions reduced the odds of healthy aging by 40%, and reporting four or more depressive symptoms reduced the odds of healthy aging by approximately 70%. After controlling for health conditions, wealth in the second quartile and early-life characteristics were no longer significantly associated with healthy aging in those without a diploma and marital status again dropped to borderline statistical significance (p = .050). In those with a high school or higher degree, the difference in the odds of healthy aging between Hispanic and non-Hispanic White adults again attained statistical significance (p = .041).

Discussion

The purpose of this investigation was to gain an understanding of the factors associated with healthy aging in the context of educational disadvantage and to examine the extent to which identified factors are similar to those associated with healthy aging in the wider, more advantaged population of midlife and older adults. Among adults without a high school diploma, age, gender, race/ethnicity, total household wealth, early-life characteristics, health practices, and the presence of physical and mental health conditions were correlated with healthy aging. With a few exceptions, the same set of factors was significantly associated with healthy aging in those with a high school or higher degree. Associations tended to be more robust to adjustment, however, in the latter educational stratum.

Consistent with existing literature (see, for example, Depp & Jeste, 2006; Hank, 2011; McLaughlin et al., 2012), advanced age reduced the odds of healthy aging in those with and without a high school diploma. Although causation cannot be inferred from this cross-sectional study, advanced age is known to pose health challenges. A notable finding is that no significant difference in the odds of healthy aging was evident for middle-aged and young-old adults among those without a high school diploma. That was not the case for individuals with a high school or higher degree. This suggests that the health advantages of middle-age may not be as great among those without a high school diploma, which may reflect earlier “wear and tear” among those aging in the context of educational disadvantage (Seeman, Epel, Gruenewald, Karlamangla, & McEwen, 2010).

Existing research examining the association of gender with healthy aging is mixed, likely owing—in part—to varying operational definitions of healthy aging. In this investigation, the odds of healthy aging were lower for women than men in both educational strata. Given that women are known to have a higher prevalence of functional health problems than men (see, for example, Crimmins, Kim, & Solé-Auró, 2011) and the definition of healthy aging used in this study emphasizes optimal functional health, this finding is not unexpected. It is noteworthy, however, that gender differences in healthy aging were not significant after controlling for health practices among those without a high school diploma. This implies that the functional health advantages of being male may not be as great among men with lower educational attainment. One contributing factor may be that men and those with lower levels of education are more likely to work in risky occupations (Baron et al., 2013) and experience work-related injuries (Pergamit & Krishnamurty, 2006) than their respective counterparts.

Consistent with reported racial inequalities in health (e.g., Nuru-Jeter, Thorpe, & Fuller-Thomson, 2011; Orsi, Margellos-Anast, & Whitman, 2010), the odds of healthy aging were lower for Black than White adults in both educational strata even after adjustment for other covariates. Hispanic ethnicity was also associated with healthy aging, but the direction of the association varied by education. Whereas the odds of healthy aging were higher for Hispanic than White adults in those without a high school diploma, they were lower among those with at least a high school degree. Given research documenting greater disability among Hispanic adults (Hayward, Hummer, Chiu, González-González, & Wong, 2014) and the use of a functional-based definition of healthy aging, the finding that Hispanic adults without a high school diploma had higher odds of healthy aging than their White counterparts seems counterintuitive. Others, however, have reported similar findings.

Using data from the National Health Interview Survey, Turra and Goldman (2007) compared mortality among Hispanic and White adults in middle and older adulthood by education. While they observed that Hispanic adults had lower death rates than White adults, the beneficial effect of Hispanic ethnicity was most pronounced among those with the least education (0 to 8 years) and reversed among those with the most (16+ years). Among other reasons, they speculated that this may reflect SES differences in health behaviors among foreign-born, Hispanic adults or differential migration patterns. Specifically, they suggested that foreign-born adults of low SES may have better health practices than their higher SES counterparts and/or be more likely to return to their country of origin when their health declines, creating a healthier population of low-SES Hispanic adults in the United States. Although neither hypothesis was tested in this study, it is noteworthy that the effect of Hispanic ethnicity was no longer significant among those without a high school diploma after controlling for health behaviors.

As reported in earlier work (Brandt, Deindl, & Hank, 2012; McLaughlin et al., 2010), individuals with lower levels of household wealth were less likely to experience healthy aging than their more advantaged counterparts in both educational strata. The health–wealth gradient was less pronounced and less robust to statistical adjustment, however, in those without a high school diploma. This may be a function of the lower level (median = $ 32,000 vs. $181,000; data not shown) and narrower range (−$765,986 to $6,688,428 vs. −$1,510,000 to $29,748,000; data not shown) of wealth accumulated by those without a high school diploma. Nevertheless, after controlling for all covariates, the odds of healthy aging were reduced by over 40% for those in the lowest quartile of total household wealth relative to those in the highest among adults who did not complete high school. While this association may partly reflect the influence of functional health problems on wealth accumulation, it is also possible that greater financial means permit access to health-promoting resources and/or reflect healthier living and working conditions (Adler & Newman, 2002; Adler & Ostrove, 1999) that make it easier to maintain health in the context of educational disadvantage.

Another sociodemographic factor that warrants brief mention is marital status. Among those without a diploma, the association between marriage/partnership and healthy aging was borderline significant in three of the four models (p ≤ .06) and statistically significant (p = .019) in another. Although this association may be entirely due to chance, it is plausible that marriage/partnership confers a protective effect in the context of educational disadvantage. Further investigation of this finding is needed.

Consistent with research documenting the imprint of childhood health and SES on health in later life (see, for example, Brandt et al., 2012; Haas, 2008; Pavela & Latham, 2015), adults growing up in a poor childhood family and those rating their childhood health as fair or poor had reduced odds of healthy aging in both educational strata. After controlling for health conditions, however, the effects of early-life characteristics were no longer significantly associated with healthy aging in those without a high school diploma. This may partly be a function of statistical power, as effect sizes were similar across educational strata in the pertinent model. Nonetheless, the effects of early-life characteristics were reduced after controlling for physical and mental health conditions, suggesting that one means by which early-life characteristics are connected to healthy aging is through their association with these conditions.

As has been reported by others, BMI (Peel, McClure, & Bartlett, 2005; Pruchno & Wilson-Genderson, 2015; Pruchno, Wilson-Genderson, Rose, & Cartwright, 2010), alcohol use (Brandt et al., 2012; Britton, Shipley, Singh-Manoux, & Marmot, 2008; Pruchno & Wilson-Genderson, 2015; Pruchno et al., 2010), and physical activity (Brandt et al., 2012; Britton et al., 2008; Depp & Jeste, 2006; Peel et al., 2005; Pruchno et al., 2010) were associated with healthy aging. Specifically, obesity reduced the odds of healthy aging by approximately 30% in those with and without a high school diploma, after controlling for health conditions. It is likely that even in the absence of chronic health conditions excess body weight makes it difficult to perform many of the tasks required to meet the definition of healthy aging (e.g., climbing several flights of stairs).

In contrast with obesity, alcohol consumption and physical activity increased the odds of healthy aging in both strata. Although little controversy exists about the benefits of physical activity, the use of alcohol as a means of promoting health is controversial (see, for example, the following discourse: Poli et al., 2013; Testino, Patussi, Leone, Scafato, & Borro, 2014). While moderate alcohol use may promote cardiovascular health (Poli et al., 2013), alcohol consumption increases the risk of some cancers (Scoccianti et al., 2015). Moreover, it should be acknowledged that part of the “effect” of alcohol observed in this cross-sectional study may stem from individuals with health problems ceasing to use alcohol. More work is needed to better understand the role of alcohol in healthy aging.

Given the known health effects of smoking, findings for ever smoking were unexpected. Specifically, ever smoking was not significantly associated with healthy aging in those without a high school diploma. One potential explanation for this finding is that dichotomizing smoking into ever versus never smoking created a group of “ever smokers” that varied widely with respect to current and former exposure to cigarette smoking, making it harder to detect a true association.

The effects of physical and mental health conditions were virtually identical across educational strata, with each additional chronic condition reducing the odds of healthy functional aging by 40% and four or more depressive symptoms reducing the odds of healthy aging by approximately 70%. Although reverse causation cannot be ruled out in the case of conditions such as depression, these findings serve as a reminder of the strong connection between physical and mental health conditions and functional health. Given that many older adults consider the ability to care for oneself to be an important element of successful aging (Phelan, Anderson, LaCroix, & Larson, 2004), the need to prevent and properly manage these conditions cannot be overstated.

Although there are some notable differences, factors associated with healthy aging among adults aging in the context of educational disadvantage are largely consistent with those observed for the wider population of midlife and older adults. Moreover, they coincide with Masten’s (2001) assertion that “ordinary” factors are responsible for positive outcomes in the face of disadvantage. Importantly, a number of the observed correlates are modifiable, which suggests that healthy aging is possible for larger numbers of adults aging in the context of educational disadvantage.

While one might conclude that healthy aging would be more common among educationally disadvantaged adults if they made better health choices, such a conclusion would be incomplete. Health-related choices are made within the confines of a set of life constraints (Adler & Newman, 2002; Bird & Rieker, 2008). For educationally disadvantaged adults, those constraints are likely to be more numerous and/or prohibitive than is the case for their more advantaged counterparts (Adler & Newman, 2002), necessitating a focus on the broader context of individuals’ lives when seeking to understand and promote healthy aging among educationally disadvantaged adults. As emphasized by Ungar (2011), positive outcomes in disadvantaged circumstances are more likely to occur in environments that facilitate them. An important next step will be to identify the environmental factors that encourage acquisition of the resources and behaviors that promote healthy aging in the context of educational disadvantage.

While this study sheds insight into the characteristics of those who experience healthy aging in the context of educational disadvantage, several limitations should be noted. First, the cross-sectional design limits causal conclusions. Second, the “educationally advantaged” subgroup includes individuals with a wide range of education. Within this broad subgroup, there may be differences in correlates of healthy aging that warrant exploration. Third, the data utilized in this investigation are self-reported and, therefore, subject to reporting error. Depending on the nature of the error, observed associations could be over- or underestimated. Fourth, measurement of some of the concepts examined in this study was suboptimal. The available measures of physical activity, for instance, contain no information on duration of engagement in the specified activities, and frequency of engagement is imprecisely assessed. Consequently, a number of those classified as “physically active” may not have achieved recommended levels of physical activity (i.e., ≥ 150 min of moderate or ≥ 75 min of vigorous aerobic exercise each week; Garber et al., 2011). As a result, the association of moderate and vigorous physical activity with healthy aging may be underestimated. Fifth, there are noteworthy limitations in the assessment of childhood health. In this investigation, childhood health was assessed using a single measure of self-rated health that has unbalanced response options. In addition, prior research suggests that this measure is less reliable among adults with lower levels of education (Haas, 2007). Notwithstanding these limitations, studies that have used this measure indicate that self-ratings of childhood health are correlated as expected with other measures of early-life health (e.g., birth weight and health conditions; Elo, 1998; Haas, 2007; Smith, 2009). Moreover, in this investigation, self-rated childhood health was associated with healthy aging in the anticipated direction in both educational strata. Sixth, institutionalized adults and those requiring proxy respondents are not represented in this investigation. To the extent that educational disadvantage increases the risk of these statuses, any bias stemming from their exclusion will be more pronounced among those without a high school diploma. Finally, this study only provides an initial look at those who experience healthy aging in the context of educational disadvantage. Our understanding would be enriched by qualitative work with adults who experience healthy aging in the context of educational disadvantage, the use of more comprehensive measures of potential determinants, and longitudinal research that incorporates elements of the social context in which aging occurs.

To conclude, nearly 19% of middle-aged and older Americans with less than a high school diploma were experiencing healthy aging in 2012. While some factors (e.g., Hispanic ethnicity) appear to operate differently across educational strata, the factors associated with healthy aging among those with less than a high school diploma are, as Masten (2001) so eloquently explained, rather ordinary. Although more work is needed, this study suggests that there are steps that can be taken to increase the number of adults who experience healthy aging in the context of educational disadvantage.

Footnotes

Author’s Note

RAND HRS Data (Version O) and corresponding RAND Enhanced Fat Files were utilized in this analysis. These data files are produced by the RAND Center for the Study of Aging (Santa Monica, CA) and are funded by the National Institute on Aging and the Social Security Administration.

Declaration of Conflicting Interests

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

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded through a grant from the University Senate Committee on Faculty Research at Miami University. HRS is sponsored by the National Institute on Aging [grant number NIA U01AG009740].

References

Adler

N. E.

Newman

(2002). Socioeconomic disparities in health: Pathways and policies. Health Affairs, 21(2), 60-76.

Adler

N. E.

Ostrove

J. M.

(1999). Socioeconomic status and health: What we know and what we don’t. Annals of the New York Academy of Sciences, 896, 3-15.

Baron

S. L.

Steege

A. L.

Marsh

S. M.

Menéndez

C. C.

Myers

J. R.

, Centers for Disease Control and Prevention. (2013). Nonfatal work-related injuries and illnesses—United States, 2010. Morbidity and Mortality Weekly Report Supplements, 62(3), 35-40.

Bird

C. E.

Rieker

P. P.

(2008). Gender and health: The effects of constrained choices and social policies. Cambridge, UK: Cambridge University Press.

Brandt

Deindl

Hank

(2012). Tracing the origins of successful aging: The role of childhood conditions and social inequality in explaining later life health. Social Science & Medicine, 74, 1418-1425. doi:10.1016/j.socscimed.2012.01.004

Braveman

Egerter

Williams

D. R.

(2011). The social determinants of health: Coming of age. Annual Review of Public Health, 32, 381-398. doi:10.1146/annurev-publhealth-031210-101218

Britton

Shipley

Singh-Manoux

Marmot

M. G.

(2008). Successful aging: The contribution of early-life and midlife risk factors. Journal of the American Geriatrics Society, 56, 1098-1105. doi:10.1111/j.1532-5415.2008.01740.x

Chien

Campbell

Chan

Hayden

Hurd

Main

. . . St. Clair

(2015). RAND HRS data documentation, version O. Santa Monica, CA: RAND Center for the Study of Aging. Retrieved from http://hrsonline.isr.umich.edu/}modules/meta/rand/randhrso/randhrs_O.pdf

Committee on Future Directions for Behavioral and Social Sciences Research at the National Institutes of Health. (2001). Positive health: Resilience, recovery, primary prevention, and health promotion. In Singer

B. H.

Ryff

C. D.

(Eds.), New horizons in health: An integrative approach (pp. 45-62). Washington, DC: National Academy Press.

10.

Crimmins

E. M.

Kim

J. K.

Langa

K. M.

Weir

D. R.

(2011). Assessment of cognition using surveys and neuropsychological assessment: The Health and Retirement Study and the Aging, Demographics, and Memory Study. The Journals of Gerontology, Series B: Psychological Sciences & Social Sciences, 66(Suppl. 1), i162-i171. doi:10.1093/geronb/gbr048

11.

Crimmins

E. M.

Kim

J. K.

Solé-Auró

(2011). Gender differences in health: Results from SHARE, ELSA and HRS. European Journal of Public Health, 21, 81-91. doi:10.1093/eurpub/ckq022

12.

Depp

C. A.

Jeste

D. V.

(2006). Definitions and predictors of successful aging: A comprehensive review of larger quantitative studies. The American Journal of Geriatric Psychiatry, 14, 6-20. doi:10.1097/01.JGP.0000192501.03069.bc

13.

Elo

I. T.

(1998). Childhood conditions and adult health: Evidence from the Health and Retirement Study (PARC Working Paper Series No. 98–03). Retrieved from http://parc.pop.upenn.edu/sites/parc.pop.upenn.edu/files/parc/PARCwps98-03.pdf

14.

Garber

C. E.

Blissmer

Deschenes

M. R.

Franklin

B. A.

Lamonte

M. J.

Lee

. . . Swain

D. P.

(2011). Quantity and quality of exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: Guidance for prescribing exercise. Medicine & Science in Sports & Exercise, 43, 1334-1359.

15.

Haas

(2007). The long-term effects of poor childhood health: An assessment and application of retrospective reports. Demography, 44, 113-135.

16.

Haas

(2008). Trajectories of functional health: The “long arm” of childhood health and socioeconomic factors. Social Science & Medicine, 66, 849-861. doi:10.1016/j.socscimed.2007.11.004

17.

Hank

(2011). How “successful” do older Europeans age? Findings from SHARE. The Journals of Gerontology, Series B: Psychological Sciences & Social Sciences, 66, 230-236. doi:10.1093/geronb/gbq089

18.

Hayward

M. D.

Hummer

R. A.

Chiu

C. T.

González-González

Wong

(2014). Does the Hispanic paradox in U.S. adult mortality extend to disability? Population Research and Policy Review, 33, 81-96. doi:10.1007/s11113-013-9312-7

19.

Health and Retirement Study. (2008). Sample evolution: 1992-1998. Retrieved from http://hrsonline.isr.umich.edu/sitedocs/surveydesign.pdf

20.

Juster

F. T.

Suzman

(1995). An overview of the Health and Retirement Study. The Journal of Human Resources, 30, S7-S56. doi:10.2307/146277

21.

Liu

Hummer

R. A.

(2008). Are educational differences in U.S. self-rated health increasing? An examination by gender and race. Social Science & Medicine, 67, 1898-1906. doi:10.1016/j.socscimed.2008.09.021

22.

Masten

A. S.

(2001). Ordinary magic: Resilience processes in development. The American Psychologist, 56, 227-238.

23.

McLaughlin

S. J.

Connell

C. M.

Heeringa

S. G.

L. W.

Roberts

J. S.

(2010). Successful aging in the United States: Prevalence estimates from a national sample of older adults. The Journals of Gerontology, Series B: Psychological Sciences & Social Sciences, 65, 216-226. doi:10.1093/geronb/gbp101

24.

McLaughlin

S. J.

Jette

A. M.

Connell

C. M.

(2012). An examination of healthy aging across a conceptual continuum: Prevalence estimates, demographic patterns, and validity. The Journals of Gerontology, Series A: Biological Sciences & Medical Sciences, 67, 783-789. doi:10.1093/gerona/glr234

25.

Meara

E. R.

Richards

Cutler

D. M.

(2008). The gap gets bigger: Changes in mortality and life expectancy, by education, 1981-2000. Health Affairs, 27, 350-360. doi:10.1377/hlthaff.27.2.350

26.

Nuru-Jeter

A. M.

Thorpe

R. J.

Fuller-Thomson

(2011). Black-white differences in self-reported disability outcomes in the U.S.: Early childhood to older adulthood. Public Health Reports, 126, 834-843.

27.

Orsi

J. M.

Margellos-Anast

Whitman

(2010). Black-White health disparities in the United States and Chicago: A 15-year progress analysis. American Journal of Public Health, 100, 349-356. doi:10.2105/AJPH.2009.165407

28.

Pavela

Latham

(2015). Childhood conditions and multimorbidity among older adults. The Journals of Gerontology, Series B: Psychological Sciences & Social Sciences. Advance online publication. doi:10.1093/geronb/gbv028

29.

Peel

N. M.

McClure

R. J.

Bartlett

H. P.

(2005). Behavioral determinants of healthy aging. American Journal of Preventive Medicine, 28, 298-304. doi:10.1016/j.amepre.2004.12.002

30.

Pergamit

M. R.

Krishnamurty

(2006, May). Multiyear nonfatal work injury rates. Monthly Labor Review, pp. 35-38. Retrieved from http://www.bls.gov/opub/mlr/2006/05/art5full.pdf

31.

Phelan

E. A.

Anderson

L. A.

LaCroix

A. Z.

Larson

E. B.

(2004). Older adults’ views of “successful aging”—How do they compare with researchers’ definitions? Journal of the American Geriatrics Society, 52, 211-216.

32.

Poli

Marangoni

Avogaro

Barba

Bellentani

Bucci

. . . Visioli

(2013). Moderate alcohol use and health: A consensus document. Nutrition, Metabolism & Cardiovascular Diseases, 23, 487-504. doi:10.1016/j.numecd.2013.02.007

33.

Pruchno

R. A.

Wilson-Genderson

(2015). A longitudinal examination of the effects of early influences and midlife characteristics on successful aging. The Journals of Gerontology, Series B: Psychological Sciences & Social Sciences, 70, 850-859. doi:10.1093/geronb/gbu046

34.

Pruchno

R. A.

Wilson-Genderson

Rose

Cartwright

(2010). Successful aging: Early influences and contemporary characteristics. The Gerontologist, 50, 821-833. doi:10.1093/geront/gnq041

35.

Raghunathan

T. E.

Solenberger

P. W.

Van Hoewyk

(2002). IVEware: Imputation and variance estimation software user guide. Retrieved from ftp://ftp.isr.umich.edu/pub/src/smp/ive/ive_user.pdf

36.

Scoccianti

Cecchini

Anderson

A. S.

Berrino

Boutron-Ruault

M. C.

Espina

. . . Romieu

(2015). European code against cancer 4th edition: Alcohol drinking and cancer. Cancer Epidemiology, 39(Suppl. 1), S67-S74. doi:10.1016/j.canep.2015.01.007

37.

Seeman

Epel

Gruenewald

Karlamangla

McEwen

B. S.

(2010). Socio-economic differentials in peripheral biology: Cumulative allostatic load. Annals of the New York Academy of Sciences, 1186, 223-239. doi:10.1111/j.1749-6632.2009.05341.x

38.

Smith

J. P.

(2009). Reconstructing childhood health histories. Demography, 46, 387-403.

39.

Steffick

D. E.

(2000). Documentation of affective functioning measures in the Health and Retirement Study. Retrieved from http://hrsonline.isr.umich.edu/sitedocs/userg/dr-005.pdf

40.

Strawbridge

W. J.

Wallhagen

M. I.

Cohen

R. D.

(2002). Successful aging and well-being: Self-rated compared with Rowe and Kahn. The Gerontologist, 42, 727-733.

41.

Taylor

M. G.

(2010). Capturing transitions and trajectories: The role of socioeconomic status in later life disability. The Journals of Gerontology, Series B: Psychological Sciences & Social Sciences, 65, 733-743. doi:10.1093/geronb/gbq018

42.

Testino

Patussi

Leone

Scafato

Borro

(2014). Moderate alcohol use and health: A nonsense. Nutrition, Metabolism & Cardiovascular Diseases, 24(1), E4-E5. doi:10.1016/j.numecd.2013.09.007

43.

Turra

C. M.

Goldman

(2007). Socioeconomic differences in mortality among U.S. adults: Insights into the Hispanic paradox. The Journals of Gerontology, Series B: Psychological Sciences & Social Sciences, 62(3), S184-S192.

44.

Ungar

(2011). The social ecology of resilience: Addressing contextual and cultural ambiguity of a nascent construct. The American Journal of Orthopsychiatry, 81, 1-17. doi:10.1111/j.1939-0025.2010.01067.x