20th Century Puerto Rico and Later-Life Health: The Association Between Multigenerational Education and Chronic Conditions in Island-Dwelling Older Adults

Abstract

Objectives:

Previous research on the association between education and older adult health in the U.S. has not included Puerto Rico. We investigated the effects of multigenerational educational attainment and chronic conditions among older Puerto Ricans residing on the archipelago’s main island.

Methods:

Data were from the longitudinal Puerto Rican Elderly Health Conditions Project. Generalized Poisson regression models were used to examine if multigenerational educational attainment was associated with chronic disease.

Results:

Findings show that parental educational attainment was associated with fewer chronic conditions among females at baseline but not at follow-up, suggesting that the effects of parental education on health over time are less pronounced. For males, educational attainment across the three generations was not significantly associated with chronic disease at baseline or follow-up.

Discussion:

Multigenerational education is an important determinant of older adult health that continues to be relevant in Puerto Rico and the Latin American and Hispanic-Caribbean region.

Keywords

Puerto Ricans Puerto Rico health disparities educational attainment social determinants of health

Introduction

The Commonwealth of Puerto Rico is an unincorporated United States (U.S.) territory with a rapidly aging population (U.S. Census Bureau, 2019). The percentage of the total population in Puerto Rico aged 65 and older has increased dramatically from 11.2% in 2000 to 21.3% in 2019 (U.S. Census Bureau, 2021). Puerto Rico currently has a larger share of adults aged 65 and older than the mainland United States (21% vs. 16.5%) (U.S. Census Bureau, 2021). This growth in the older adult population primarily stems from the outmigration of younger cohorts fueled by the financial crises associated with the $70 billion public debt and Hurricanes María and Irma (Flores & Krogstad, 2019). Consequently, the growing number of older adults is invariably going to be associated with increased rates of morbidity and disability (Coleman, 2001), which will place considerable strains on an already fragile health care system (Mulligan, 2014). Older adults in Puerto Rico have a high prevalence of cardiovascular diseases, cancer, and diabetes (Rodriguez-Ayuso et al., 2012), and compared to Latinos living in the contiguous U.S., Puerto Ricans are more likely to have hypertension, diabetes, and lung disease (Pérez & Ailshire, 2017).

Several studies have examined early life exposures (e.g., poor nutrition, exposure to infectious diseases, and socioeconomic conditions) to understand the etiology of chronic conditions among this population (McEniry, 2011; McEniry et al., 2008; McEniry & Palloni, 2010); however, chronic diseases have multiple causative factors across the life course, which includes critical exposures in early life and the accumulation of detrimental exposures throughout the life course. In addition, socioeconomic conditions in different stages of the life course have been shown to have a robust association with health in older adulthood, with persons of low socioeconomic status in early-, mid-, and late-life exhibiting poorer health in older adulthood (Poulton et al., 2002).

Currently, few studies have explored whether the relationship between education, particularly family educational attainment, and health outcomes in Latin American countries and the U.S. operate similarly. In the case of Mexico, Yahirun et al. (2017) found a marginal negative association between children’s educational attainment and parents’ functional decline. In contrast, offspring’s educational attainment has a robust protective effect on parents’ cognitive health in the U.S. (Yahirun et al., 2020). Like other Latin American countries, the positive relationship between education and health in Mexico may not confer similar benefits that have been observed in the U.S. due to educational reforms and economic transitions that occurred in the 20^th century (Ma et al., 2021). Given Puerto Rico’s economic and political relationship with the U.S., the association between educational attainment and health may or may not operate as observed in these two contexts. Hence, there is a particular need for information on the socioeconomic determinants of health among island Puerto Ricans since there have been shifts in educational opportunities that are different from that in the mainland U.S., which are relevant to later life health among this population. This study examines the relationship between various indicators of educational attainment and chronic conditions among older island-dwelling Puerto Rican adults by sex. We aim to assess the role of parental education, respondent’s education, and children’s education as determinants of later life health.

Background

Puerto Rico in the 20^th and 21^st Century

In 1898, Puerto Rico experienced a political transition from a Spanish province to a U.S. military colony. During the first years of the 20^th century, the U.S. government put initiatives to restructure the island’s educational system, mainly centered on personal hygiene and healthy behaviors (Del Moral, 2018). Even though this initiative increased accessibility to primary education among those who could not afford it and/or lived in areas distant from San Juan, illiteracy rates prevailed. In 1904, only 16% of school-age children were enrolled in any of the 1060 schools around the island (López Yustos, 1991). This trend did not change significantly until the 1950s (Rivera-Batiz & Santiago, 1996). Historians and social scientists attribute the reconstruction’s failures to governmental stakeholders’ unawareness of the Puerto Rican social reality (Rivera Median & Ramírez, 1985).

During the first half of the 20^th century, Puerto Ricans—especially those with little to no formal education—did not view education as a major priority because of the focus on a largely present agrarian economy. These educational attitudes partially explain low school enrollment during this period among rural communities where household income was based on agricultural work (Esterrich, 2018). For instance, young and adolescent males mainly joined the workforce alongside the head of the household in plantation fields. In contrast, their female counterparts stayed at home and helped their mothers with domestic diligences (e.g., cooking, cleaning, and childcare) (Wheleham, 1995). As a result, most young children dropped out early or did not enroll in school even when residing near a school. Like other agricultural societies, earning opportunities to support a Puerto Rican household drove decision-making around children’s educational possibilities (López Yustos, 1991). In particular, young males were more likely to miss school or drop out to earn a salary and support their families while facing economic hardship within an agrarian economy. Meanwhile, parents with the highest level of education were more likely to prioritize their children’s education and have a stable financial situation; thus, reducing their children’s likelihood of dropping out of school (Esterrich, 2018). Moreover, parents would transfer their knowledge about personal hygiene and health behaviors to their children.

During the late 1930s and throughout the 1940s, the local government, with the help of U.S. federal funding related to the New Deal, rapidly enacted multiple bills and implemented economic incentive projects to revitalize Puerto Rico’s national economy. These political measures contributed to a larger industrialization project known as Operación Manos a la Obra—Operation Bootstrap (Dietz, 1986). Although this project boosted Puerto Rico’s public economy, it exacerbated class inequality between people who had some formal education and those who did not; and individuals who resided or migrated to urban areas versus those who stayed in rural communities (Pantojas-García, 1990). Children of parents with low education, particularly illiterate parents from rural areas, suffered the consequences of the class struggle through resource scarcity and fewer chances of attending school.

When Puerto Rico became a U.S. Commonwealth in 1953, primary education became an obligatory mandate with its own constitution, drafted almost identically to the U.S. Constitution (López Yustos, 1991). For three decades, the local government invested in its public education system and the University of Puerto Rico, which increased school-aged children’s enrollment up to 800,000 by the end of 1970 (Rivera-Batiz & Santiago, 1996). However, by the end of the 20^th century, the public school system experienced dramatic budgetary shifts. First, school enrollment declined in the 1980s, prompted the reduction of funds, and affected the quality of education because of lower teacher hiring. Then, in August of 1990, the enactment of the Ley Orgánica increased education spending to compete with the global market and boost the island’s economy. To reach this goal, the Department of Education executed major curricular reforms focused on high-tech skills required by the global economy of the forthcoming century (ELA Puerto Rico, 1990). In addition, this bill contributed to the improvement of the University of Puerto Rico, consequently prompting an increase of college-educated locals from 70,283 to 107,551 (Bauman & Graft, 2003).

Education and Linked Lives

Education is a well-established determinant of health, with the general pattern of better health among those with higher educational attainment (Hummer & Lariscy, 2011; Jackson, 2009; Mirowsky & Ross, 2003). Specifically, individuals with lower levels of education are more likely to have cancer, lung disease, diabetes, and cardiovascular disease (Assari et al., 2020; Harvie et al., 2015; Menvielle et al., 2009; Telfair & Shelton, 2012).

Educational attainment plays an important role in health through several interrelated pathways. First, higher educational attainment can protect against risk factors for poor health due to access to health-promoting resources (e.g., healthy foods) and effective use of health-promoting information (Ross & Mirowsky, 2010). For example, individuals with a college education are more likely to have better employment opportunities, which is strongly associated with increased access to economic (e.g., income) and material resources (e.g., health insurance, healthy foods) that promote better health (Mirowsky & Ross, 1998, 2003; Phelan & Link, 2013). Moreover, educational attainment influences psychosocial factors such as greater perceived personal control (Seeman & Seeman, 1983), which encourages the adoption of healthy behaviors, including the reduced likelihood of smoking and moderate alcohol consumption, better healthy eating habits, and increased recreational physical activity, which decrease the risk for poor health and chronic conditions (Brunello et al., 2016). There is also evidence that educational attainment is associated with environmental determinants of health. For example, individuals with higher levels of education are more likely to live in residential environments that have physical activity resources (e.g., equipment, nearby parks), access to healthy foods, and other built environment characteristics related to the diet and physical activity of residents (Cerin & Leslie, 2008; Creatore et al., 2016; Van Cauwenberg et al., 2016; Wolfson et al., 2019).

Beyond personal educational attainment, individuals can benefit from others’ achievements and experiences as well. The linked lives perspective proposes other pathways by pinpointing the shared benefits and adverse effects of family members’ experiences and socioeconomic determinants (e.g., education, income, occupation) on other individuals’ health across the life course (Elder et al., 2003). Family health scholars have applied the linked lives perspective to explain the association between a family event and children’s health and behavioral outcomes. For example, data from the National Survey of Midlife Development has shown how exposure to multiple adverse childhood events—including parental job loss, alcohol/drug abuse, divorce, death—elevated young to older adults’ risk of diabetes, obesity, and heart disease (Friedman et al., 2015).

Other researchers have also used the linked lives perspective to determine the relationship between one family member’s socioeconomic status (SES) (e.g., father’s occupation, children’s educational attainment) and other member’s health (Greenfield & Moorman, 2019; Wolfe et al., 2018). Parental level factors, such as unemployment and educational attainment, influence their children’s health through resource availability. For example, parents with little to no schooling are less likely to secure employment and regularly contribute to their household income; thus, increasing their chances of experiencing economic deprivation (Schnittker, 2004). Hence, parental education influences access to basic resources associated with better childhood health (e.g., healthy foods, clean water, health care) and adulthood socioeconomic status (e.g., school materials). Moreover, parental-level factors are more likely to trigger family conflict or strain, leading to detrimental physiological and mental health consequences (Thomas, 2016). Based on the allostatic load perspective, constant stress is associated with adverse physical health outcomes due to the human body’s natural response to distress that is harmful when it occurs sporadically (McEwen, 1998). Thus, children who grow up in a stressful household environment are more likely to develop comorbidities—notably cardiovascular diseases—in later life.

More recent research examining the role of multigenerational educational attainment considers the “upstream” effect of children’s education shaping parents’ health (Friedman & Mare, 2014; Zimmer et al., 2016). Highly educated children are more likely to impart knowledge to their older parents by helping them navigate and encourage the use of health care services (Jiang & Kaushal, 2020) and prompt them to embrace healthier behaviors. For example, parents of more educated children are more likely to quit smoking (Friedman & Mare, 2014) and drink moderately (Mangiavacchi & Piccoli, 2018). Furthermore, well-educated children have greater access to socioeconomic resources (Mirowsky & Ross, 2003) they can use to provide direct care to their parents (Torssander, 2013) or pay for caregiving services (Jiang & Kaushal, 2020). With data from the National Longitudinal Survey of Mature Women, Wolfe and colleagues (2018) identified children’s educational attainment had a protective effect on silent generation mothers’ mortality.

Although we focus on education as a crucial determinant of health, we want to acknowledge that education is not the only socioeconomic status measure that influences health. Other socioeconomic measures, such as income and occupational status, are associated with health care, nutrition, housing, and schooling (Adler & Newman, 2002). However, evidence supports that education does a better job in some settings at capturing human capital factors (Ross & Mirowsky, 2010) that shape health outcomes and lifestyles (Cockerham, 2013) across the life course (Ross & Wu, 1996) relative to other socioeconomic measures. Moreover, unlike income and occupational prestige, educational attainment is less likely to be affected by illness in adulthood and does not decrease over time (Shavers, 2007).

While there is substantial evidence of the positive relationship between educational attainment and health (Hummer & Lariscy, 2011; Jackson, 2009; Mirowsky & Ross, 2003), these findings are primarily based on data from high-income countries (e.g., U.S.). In addition, the extant research examining the health benefits of family education in low- and middle-income countries presents mixed findings across Latin American and Asian countries (Smith-Greenaway et al., 2018; Yang et al., 2016; Zimmer et al., 2002, 2007). For example, like other low- and middle-income countries, the Puerto Rican public education system went through significant changes that affected educational opportunities for young Puerto Ricans and their families during the first half of the 20^th century (López Yustos, 1991; Rivera-Batiz & Santiago, 1996). Moreover, the economic circumstances described above may have altered the social meaning of educational attainment (Shavers, 2007).

The Gendered Life Course

A gendered life course framing emphasizes how men and women have distinct life trajectories due to cultural practices and policies that reproduce gender inequality in many facets of life (Moen, 2001). Importantly, this approach recognizes how individual health in late life is the accumulation of exposures experienced over a lifetime. In the case of gendered experiences in Puerto Rico, one must emphasize the historical context in which individual lives have been shaped. For instance, respondents in the Puerto Rican Elderly Health Conditions Project (PREHCO) were primarily born during the 1^st half of the 20^th century, a period where both men and women raised families at a time of considerable hardship associated with profound economic and social changes on the island. This meant that the respondents of PREHCO during this period did not have equal opportunities to obtain an education, with males more likely to have attended school (Rivera-Batiz & Santiago, 1996). Previous studies have documented low levels of education among the PREHCO sample (Pérez & Ailshire, 2017), which affects opportunities for employment and health.

The social organization of work typically structures men, particularly those with low education, to work in hazardous occupations (e.g., construction work, truck driving, farming, industrial). These occupations are associated with increased exposure to pollutants, physical hazards, strenuous physical labor, occupational-related stressors, and higher odds of fatalistic accidents (Cockerham, 2017). In addition, the conditions surrounding these types of occupations (e.g., prolonged repetitions, heavy lifting, static postures, and insufficient recovery time) are associated with an increased likelihood of developing cardiovascular illnesses and mental health problems (Coenen et al., 2018; Holtermann et al., 2012; Holtermann et al., 2018). Moreover, men are more likely to engage in hazardous health behaviors, such as binge drinking (Elliott, 2013), smoking (Syamlal et al., 2014), and participating in dangerous activities (e.g., speeding) (Cockerham, 2006) that lead to poorer health outcomes. In contrast, women are more likely to engage in healthier behaviors, including healthy eating (Mróz et al., 2011; Rothgerber, 2013), visiting their primary care physician annually, and following medical recommendations (Courtenay, 2000). However, women’s domestic responsibilities—such as parenting, homemaking, and caregiving—may intervene with these behaviors, especially among women in the lower socioeconomic stratum (Rieker et al., 2010).

Study Purpose

Considering the historical, economic, and social context of Puerto Rico, the aim of this study was to investigate the linkages between multigenerational educational attainment and chronic conditions in later life by sex. The findings from this investigation can provide information on how multigenerational structures shape resources across the life course in a setting where there have been transformative social and economic changes in Puerto Rico.

Methods

Data

We used data from the baseline (Wave 1) and four-year follow-up (Wave 2) of the Puerto Rican Elderly Health Conditions Project (PREHCO) (Palloni et al., 2013). PREHCO is a representative longitudinal study of the non-institutionalized population aged 60 and older in Puerto Rico, including respondents born between 1896 and 1943 at baseline. The sample is a multistage, stratified sample of older adults residing in Puerto Rico, with oversamples of regions heavily populated by people of African descent and individuals older than age 80. The resident populations of the island municipalities of Culebra and Vieques were excluded from the study. 4291 face-to-face interviews were conducted in Spanish between May 2002 and May 2003, with an overall response rate of 93.9%. The four-year follow-up of PREHCO included 3891 interviews conducted between June 2006 and November 2007, with a response rate of 90.7%. Information on the overall design, sampling procedures, and survey instruments of PREHCO has been previously described (McEniry & Palloni, 2010; Palloni et al., 2013). We want to note that although the PREHCO baseline data were collected 20 years ago, these data provide relevant benchmarks to better understand education and older adult health in the Latin America and the Hispanic-Caribbean region.

Our main analytical sample included 3238 respondents (1991 females; 1247 males) in Wave 1 who were directly interviewed. We excluded 578 respondents from the PREHCO sample who had completed their interviews with a proxy since proxies were not asked all interview questions relevant to the current analysis. An additional 390 respondents were excluded since they reported having no children. Lastly, 85 respondents were excluded since they did not report the sex of their child(ren). Moreover, approximately 50% of the analytical sample had missing data on at least one of the independent variables or covariates, with item non-response highest on parental education (34% missing on father’s education, 26% missing on mother’s education). We had very low missingness (if any) on the other variables included in the analysis. For example, the indicator on leaving school to work had 4% missingness (the next highest non-response item), and experiencing economic deprivation in childhood had 1% missingness. Almost all the missingness in our analytical sample is due to parental education. To address the missing data in our sample, we employed multiple imputation (details provided below).

Between the baseline and follow-up interviews, 360 respondents died, and 294 were lost at follow-up, resulting in a sample of 2584 individuals available for Wave 2 (of the 3238 respondents in the analytical sample in Wave 1). Additionally, we excluded 249 respondents from the Wave 2 sample who had completed interviews with a proxy since proxies were not asked the health questions relevant to the analysis. Thus, our analytical sample for the Wave 2 analysis included 2335 individuals (1470 females; 865 males).

Measures

Chronic Disease

The presence of chronic disease at baseline and the four-year follow-up was assessed using the following question, “Has a doctor ever told you if you have [condition]?” Respondents reported whether they had any of the following six conditions: high blood pressure or hypertension, heart disease, diabetes, stroke, lung disease, and cancer. Respondents who reported taking medications to decrease high blood pressure and manage diabetes were coded as having hypertension and diabetes, respectively. We created a chronic disease scale that top-coded the number of chronic conditions at three or more conditions (i.e., 0 = no chronic conditions, 1 = 1 chronic condition, 2= 2 chronic conditions, and 3 = 3+ chronic conditions). Top-coding was used to deal with the limited number of individuals who reported more than three chronic conditions at baseline and follow-up and to improve the statistical performance of the variable (Dye & Mitchel, 2010).

Educational Attainment

Educational attainment was operationalized within the historical context of each generation generated from meaningful educational thresholds. Parental educational attainment was categorized based on the responses provided by PREHCO, which included (1) no education, (2) less than 8^th-grade education, and (3) 8^th-grade education or more. For most respondents, a high proportion of their parents did not complete more than an 8^th-grade education than did future generations (see Table 1). This is not surprising given that the respondent’s parents grew up when access to education was considered a private entity in an agrarian society. As education was made more accessible, the average years of schooling increased. To reflect this trend, we categorized respondents’ education based on the education system Puerto Rico implemented in 1899 that included mandatory primary schooling and the School Laws that followed (Bobonis & Toro, 2007): (1) 0–5 years of education, which reflects not completing mandatory primary or elementary education; (2) 6–11 years of education, reflecting individuals who completed primary or elementary education but did not complete secondary education; and (3) 12 or more years of education, reflecting individuals who completed secondary education. For adult children, we used the educational attainment of the highest educated child, an approach that has been used previously (Torssander, 2013; Wolfe et al., 2018) and was operationalized as (1) 0–12 years of education, reflecting primary and secondary education, (2) 13–16 years of education, reflecting tertiary or higher education, and (3) 17 or more years of education, reflecting advanced or professional education.

Table 1.

Baseline Sample Characteristics of Puerto Ricans Aged 60+ by Sex, 2002–2003 and 2006–2007.

	Wave 1				Wave 2
	Female		Male		Female		Male
	% or Mean	SD	% or Mean	SD	% or Mean	SD	% or Mean	SD
Number of chronic conditions	1.3	1.8	1.2	1.4	1.2	1.0	1.1	1.3
0 conditions	24.1		27.7		23.2		28.7
1 condition	35.9		36.7		43.4		39.1
2 conditions	26.7		25.8		26.1		23.8
3+ conditions	13.3		9.8		7.2		8.4
Multigenerational educational attainment
Father’s education
No education	31.7		32.9		30.7		31.8
<8th grade	47.6		47.1		48.6		48.1
8th Grade	20.8		20.0		20.7		20.1
Mother’s education
No education	39.8		35.5		38.7		31.4
<8th grade	44.4		48.6		44.9		51.7
8th Grade	16.0		15.9		16.5		16.9
Respondent education
0–5 years	33.5		25.8		30.3		22.8
6–11 years	32.0		34.7		33.9		37.4
12 or more years	34.4		39.5		35.8		39.7
Child’s education
0–12 years	38.0		36.7		35.5		34.0
13–16 years	47.7		47.1		49.9		49.4
17+ years	14.3		16.1		14.6		16.6
Respondent childhood experiences
Economic deprivation	45.5		50.0		46.5		50.5
Left school to work	31.7		41.8		32.0		41.0
Demographics
Age	69.9	12.4	69.8	9.3	68.8	10.9	68.6	7.8
Birth cohort
1901–1915	3.4		3.0		1.5		1.0
1916–1929	29.7		29.8		26.1		25.8
1930–1943	67.0		67.2		72.4		73.2
Residential environment before age 18
Countryside	51.7		59.1		51.7		58.2
City/Suburbs/Elsewhere	48.3		40.9		48.3		41.8
Number of siblings	7.0	5.9	6.9	5.5	7.1	5.8	6.9	5.8
Number of children	4.0	4.8	4.1	4.3	3.9	4.2	4.1	3.8
Sex of highest educated child male	48.0		57.2		48.1		55.4
Sex of highest educated child female	52.0		42.8		51.9		44.6
Unweighted N	1991		1247		1470		865

Note. All variables are weighted using PREHCO sample weights; SD = standard deviation.

Childhood Experiences

We include three measures of childhood experiences that may affect a respondent’s access to higher education. Childhood economic deprivation is a dichotomous variable and is based on whether the respondent suffered financial hardship that prevented them from eating regularly, dressing appropriately, or receiving necessary medical attention. Leaving school to work is a dichotomous variable and was based on whether the respondent or any of their siblings had to drop out of school during their childhood or adolescence and work to help their parents. We would like to note that a respondent who faced economic deprivation in childhood may reflect their father’s socioeconomic position (SEP) more so than their mother’s SEP. However, a mother’s SEP may be influential for early-life nutrition. The type of residential environment that a respondent lived in before they were 18 is a dichotomous variable that indicated whether they lived in the countryside or the city/suburbs/elsewhere. We designated the suburban and “elsewhere” response categories as part of the same category as city because these places are adjacent to the city and represented a small proportion of respondents (<5%).

Sociodemographics

Every model in the analysis controls for several important factors related to multigenerational educational attainment and chronic disease (see Table 1). Age is a continuous variable that ranges from age 60 to age 102, including respondents born between 1901 and 1943. Three birth cohorts of older Puerto Ricans are included that reflect 14- or 15-year intervals: (1) 1901–1915, (2) 1916–1929, and (3) 1930–1943. The number of siblings is a continuous variable that reflects how many siblings the respondent has/had in their lifetime, ranging from 0 to 20 siblings. This variable is included because there may be a potential confounder in a respondent’s educational attainment based on the available resources in a (large) family to support all children attending school. The number of children is a continuous variable that reflects how many children (i.e., live births) the respondent has/had in their lifetime, ranging from 1 to 20 children. The sex of the respondent’s most educated child is a dichotomous variable indicating whether they are female or male.

Analytic Strategy

To address item non-response, we imputed data on the independent variables and covariates at baseline using multiple chained equations with the mi impute command in Stata V.16.1 (StataCorp, 2019). Imputation models included all analytic variables as well as variables not included in our analysis that were theoretically related to item non-response or the analytical variables (e.g., parent’s ability to read or write to predict parental education) (Heeringa et al., 2017). In addition, we followed the imputation guidelines recommended by Graham and colleagues (Graham et al., 2007, 2013).

The primary goal of this analysis was to assess how educational attainment across generations affects the physical health of respondents in later life by sex. We estimated multivariable generalized linear models (GLM) with a Poisson distribution and log link function (or, Poisson GLM) to calculate rate ratios to assess the association between parent, personal, and adult children’s educational attainment and chronic disease separately before fitting a model that included the educational attainments of all three generations. These analyses were stratified by sex. The Poisson GLM modeling strategy was chosen since the model accounts for the underdispersion observed in the dependent variable (i.e., the number of chronic conditions) (Consul & Famoye, 1992; Famoye, 2015; Harris et al., 2012). Lastly, the same set of analyses was conducted using the number of chronic conditions at follow-up (i.e., Wave 2) as the dependent variable to determine if any of the educational attainment and health differences observed at baseline hold over time among surviving respondents. We included the number of chronic conditions reported at Wave 1 across all models.

All statistical analyses for the present study were conducted in Stata V.16.1 (StataCorp, 2019). In all analyses, the complex sampling design of PREHCO was accounted for using Stata’s svy suite of commands.

Results

Baseline Sample Characteristics

Table 1 presents the characteristics of the 3238 respondents (1991 females; 1247 males) included in our main analysis by sex. On average, females reported a slightly higher number of chronic conditions (x̄=1.3; σ=1.8) relative to their male counterparts (x̄=1.2; σ=1.4). Nevertheless, female respondents were more likely to report three or more chronic conditions (13.3%) than male respondents (9.8%). Regardless of respondent sex, a higher proportion of respondents reported that their fathers and mothers had primarily a less than 8^th-grade education. In addition, regardless of sex, most respondents reported a less than high school education. However, respondents had greater educational gains than their fathers and mothers. This is also true of respondents’ children, who the majority were able to obtain some post-secondary education or more. Male respondents were more likely to have experienced childhood adversity, with males more likely to report having experienced economic deprivation (50.0% vs. 45.5%, respectively) and more likely to (or have a sibling) drop out of school to work (41.8% vs. 31.7%, respectively).

Table 2 presents the bivariate associations between multigenerational educational attainment and the number of chronic conditions by sex at baseline to demonstrate the direct relationship of these variables. Respondents who had parents who completed at least an 8th-grade education reported fewer chronic conditions than respondents who had parents who had less than an 8^th-grade education. However, female respondents were more likely to report a higher number of chronic conditions (x̄=1.46; σ=3.45) relative to their male counterparts (x̄=1.21; σ=2.33) whose parents had no education. Respondents who achieved higher levels of education also reported a lower number of chronic conditions. Although, female respondents who reported completing 0–5 years of education were more likely to report more chronic conditions (x̄=1.51; σ=2.32) than their male counterparts (x̄=1.16; σ=3.17). Respondent’s children’s educational attainment had different associations with the number of chronic conditions reported by female and male respondents. Among females, their children’s higher levels of educational attainment were associated with a lower number of chronic conditions reported. In contrast, among males, their children’s higher levels of educational attainment were associated with a higher number of chronic conditions reported.

Table 2.

Bivariate Associations Between Multigenerational Educational Attainment and Chronic Conditions of Puerto Ricans Aged 60+ by Sex (n=3238).

	Female			Male
	Mean	SD	95% CI	Mean	SD	95% CI
Parental educational attainment
Father’s education
No education	1.46	3.45	[1.31, 1.61]	1.21	2.33	[1.08. 1.34]
<8th grade	1.30	2.27	[1.20, 1.40]	1.19	2.20	[1.07, 1.32]
8th Grade	1.17	3.61	[1.01, 1.33]	1.17	3.00	[0.99, 1.33]
Mother’s education
No education	1.46	3.60	[1.30, 1.62]	1.25	2.16	[1.13, 1.37]
<8th grade	1.26	2.23	[1.17, 1.36]	1.17	2.12	[1.05, 1.29]
8th Grade	1.13	3.71	[0.97, 1.30]	1.15	3.21	[0.97, 1.33]
Respondent educational attainment
0–5 years	1.51	2.32	[1.38, 1.64]	1.16	3.17	[1.02, 1.30]
6–11 years	1.31	2.45	[1.17, 1.44]	1.22	3.44	[1.07, 1.37]
12 or more years	1.15	1.47	[1.07, 1.23]	1.20	2.43	[1.09, 1.31]
Children’s educational attainment
0–12 years	1.32	2.65	[1.21, 1.44]	1.10	2.09	[0.98, 1.22]
13–16 years	1.35	2.23	[1.25, 1.45]	1.23	2.25	[1.11, 1.36]
17+ years	1.23	3.33	[1.09, 1.38]	1.29	3.05	[1.12, 1.46]
Unweighted N	1991			1247

Note. All variables are weighted using PREHCO sample weights; SD = standard deviation; CI = confidence intervals.

Educational Attainment and Chronic Conditions at Baseline

Table 3 presents results from Poisson GLMs for Puerto Rican females at baseline. Rate ratios (RRs) and their 95% confidence intervals (CIs) are shown. The RRs are interpreted similarly to odds ratios, with RRs between zero and one indicating a reduction in the average number of chronic conditions reported. RRs greater than one indicate an increase in the average number of chronic conditions reported. We found that the educational attainment of fathers, mothers, and that of the respondent was initially associated with the average number of chronic conditions reported among female Puerto Ricans. In Model 1, we found that fathers with at least an 8^th-grade education was associated with a 19% reduction (RR = 0.81, 95% CI [0.69, 0.95]) in the average number of chronic conditions reported among females relative to respondents whose father had no education. In Model 2, mothers with at least an 8^th-grade education was associated with a 20% reduction (RR = 0.80, 95% CI [0.67, 0.96]) in the average number of chronic conditions reported relative to respondents whose mother had no education. Model 3 revealed that a female respondent with at least 12 years of education was associated with a 14% reduction (RR = 0.86, 95% CI [0.76, 0.98]) in average the number of chronic conditions reported relative to respondents with less than six years of education. In addition, females that did not live in the countryside (i.e., lived in the city/suburbs/elsewhere) before the age of 18 was associated with a 13% reduction in the average number of chronic conditions reported. However, Model 4 showed that a respondent’s child’s educational attainment was not related to the average number of chronic conditions reported. In Model 5, when all three generations of educational attainment were included, the significant associations observed with parental and personal educational attainment were reduced to non-significance. This may be due to the multicollinearity of educational attainment across the three generations (see Appendix A). In addition, across the models, increasing age and later-born cohorts among female respondents were associated with an increase in the average number of chronic conditions reported.

Table 3.

Rate Ratios (RR) With 95% Confidence Intervals (95% CI) for Chronic Conditions Among Puerto Rican Females Aged 60+, 2002–2003 (n=1991).

	Model 1			Model 2			Model 3			Model 4			Model 5
	RR		95% CI	RR		95% CI	RR		95% CI	RR		95% CI	RR		95% CI
Multigenerational educational attainment
Father’s education (ref = no education)
<8th grade	0.90		[0.81, 1.01]										0.97		[0.86, 1.09]
8th Grade	0.81	*	[0.69, 0.95]										0.94		[0.78, 1.13]
Mother’s education (ref = no education)
<8th grade				0.89		[0.78, 1.02]							0.96		[0.84, 1.09]
8th Grade				0.80	*	[0.67, 0.96]							0.96		[0.78, 1.18]
Respondent education (ref = 0–5 years)
6–11 years							0.92		[0.83, 1.02]				0.93		[0.84, 1.04]
12 or more years							0.86	*	[0.76, 0.98]				0.89		[0.78, 1.01]
Child’s education (ref = 0–12 years)
13–16 years										1.03		[0.94, 1.13]	1.08		[0.99, 1.18]
17+ years										0.95		[0.83, 1.09]	1.06		[0.92, 1.22]
Respondent childhood experiences
Economic deprivation							1.01		[0.91, 1.12]				1.00		[0.90, 1.11]
Left school to work							1.02		[0.91, 1.15]				1.02		[0.91, 1.14]
Demographics
Age	1.01	*	[1.00, 1.03]	1.01		[1.00, 1.03]	1.02	**	[1.00, 1.03]	1.01	*	[1.00, 1.03]	1.01	*	[1.00, 1.03]
Birth cohort (ref = 1901–1915)
1916–1929	1.37	*	[1.04, 1.79]	1.40	*	[1.07, 1.84]	1.41	*	[1.08, 1.85]	1.42	*	[1.08, 1.88]	1.41	*	[1.08, 1.84]
1930–1943	1.40		[0.98, 2.01]	1.46	*	[1.02, 2.11]	1.48	*	[1.05, 2.08]	1.48	*	[1.03, 2.12]	1.46	*	[1.04, 2.07]
Lived in the City/Suburbs/Elsewhere before age 18							0.87	*	[0.78, 0.98]				0.89	*	[0.79, 1.00]
Number of siblings							1.01	*	[1.00, 1.02]				1.01		[1.00, 1.02]
Number of children										1.03	***	[1.01, 1.04]	1.02		[1.00, 1.03]
Sex of highest educated child is female										0.99		[0.89, 1.10]	0.99		[0.90, 1.09]

Note. All variables are weighted using PREHCO sample weights; RR = Rate ratio; CI = Confidence interval; * p < .05; ** p < .01; *** p < .001.

In contrast to Puerto Rican females, males’ parental educational attainment was not significantly associated with the average number of chronic conditions reported (see Table 4). The association between a male respondent’s educational attainment and chronic conditions was also not significant. However, male respondents who experienced economic deprivation in childhood were associated with an increase in the average number of chronic conditions reported (RR = 1.23, 95% CI [1.10, 1.38]) in older adulthood than males who did not experience any economic deprivation. Children’s educational attainment was also not associated with chronic conditions. When all three generations of educational attainment were included in the final model, none were associated with the number of chronic conditions reported. However, males who experienced economic deprivation in childhood were still associated with an increase in the average number of chronic conditions reported (RR = 1.24, 95% CI [1.10, 1.39]). Like females, Puerto Rican males who were older, were born in later cohorts, and had more siblings reported a higher number of chronic conditions across the models.

Table 4.

Rate Ratios (RR) With 95% Confidence Intervals (95% CI) for Chronic Conditions Among Puerto Rican Males Aged 60+, 2002–2003 (n=1247).

	Model 1			Model 2			Model 3			Model 4			Model 5
	RR		95% CI	RR		95% CI	RR		95% CI	RR		95% CI	RR		95% CI
Multigenerational educational attainment
Father’s education (ref = no education)
<8th grade	1.01		[0.87, 1.16]										1.04		[0.91, 1.20]
8th Grade	0.99		[0.84, 1.18]										1.09		[0.90, 1.33]
Mother’s education (ref = no education)
<8th grade				0.95		[0.84, 1.08]							0.96		[0.84, 1.09]
8th Grade				0.96		[0.80, 1.14]							0.97		[0.79, 1.21]
Respondent education (ref = 0–5 years)
6–11 years							1.08		[0.92, 1.27]				1.06		[0.90, 1.24]
12 or more years							1.14		[0.96, 1.34]				1.07		[0.90, 1.29]
Child’s education (ref = 0–12 years)
13–16 years										1.11		[0.97, 1.27]	1.09		[0.95, 1.26]
17+ years										1.19	*	[1.00, 1.41]	1.16		[0.97, 1.40]
Respondent childhood experiences
Economic deprivation							1.23	***	[1.10, 1.38]				1.24	***	[1.10, 1.39]
Left school to work							0.93		[0.82, 1.05]				0.93		[0.82, 1.05]
Demographics
Age	1.02	*	[1.00, 1.03]	1.02	*	[1.00, 1.03]	1.02	*	[1.00, 1.03]	1.02	*	[1.00, 1.03]	1.02	*	[1.00, 1.03]
Birth cohort (ref = 1901–1915)
1916–1929	1.58	*	[1.11, 2.24]	1.58	*	[1.12, 2.25]	1.51	*	[1.08, 2.11]	1.56	*	[1.11, 2.20]	1.51	*	[1.10, 2.08]
1930–1943	1.77	*	[1.13, 2.78]	1.78	*	[1.13, 2.80]	1.66	*	[1.06, 2.60]	1.78	*	[1.15, 2.76]	1.69	*	[1.10, 2.59]
Lived in urban environment before age 18							0.94		[0.83, 1.07]				0.93		[0.83, 1.05]
Number of siblings							1.02	*	[1.00, 1.03]				1.02	*	[1.00, 1.03]
Number of children										1.01		[0.99, 1.03]	1.01		[0.98, 1.03]
Sex of highest educated child female										0.85	**	[0.75, 0.96]	0.84	**	[0.74, 0.95]

Note. All variables are weighted using PREHCO sample weights; RR = Rate ratio; CI = Confidence interval; * p<.05; ** p<.01; *** p<.001.

Educational Attainment and Chronic Conditions at Follow-Up

The results from the longitudinal analysis using baseline covariates to predict the number of chronic conditions reported at the four-year follow-up among surviving PREHCO respondents revealed no association between multigenerational educational attainment and chronic conditions among female (Appendix B) and male (Appendix C) respondents. However, ancillary analyses comparing surviving respondents versus respondents not included in Wave 2 revealed that those who reported a lower average number of chronic conditions, had higher levels of education, had a child that achieved more than a bachelor’s education, and were younger at baseline were more likely to have survived (results available upon request). This suggests that the surviving respondents of PREHCO in the four-year follow-up are selected survivors. Namely, these survivors had more advantageous demographic profiles at baseline relative to those who attrited. Moreover, given that parental education was not associated with survival, but higher levels of respondent education were, this may suggest that the effects of parental education on survival and health at follow-up are less pronounced since well-educated individuals accumulate resources as they age and are less dependent on any one resource over time (Ross & Mirowsky, 2011). Thus, there are structural processes underlying the health status of older Puerto Ricans surviving to older ages that need to be further examined.

Discussion

The older adult population in the Commonwealth of Puerto Rico is growing rapidly (U.S. Census Bureau, 2019). This demographic trend will be associated with greater rates of morbidity and disability in the population, consequently adding more pressure to the archipelago’s unstable health care system (Mulligan, 2014). In order to prevent or delay the occurrence of age-related diseases, reduce health disparities, and improve the well-being of current and future cohorts of older Puerto Ricans, it is imperative to understand the socioeconomic conditions and unequal opportunities that exist in Puerto Rico. Well-established scholarship demonstrates a protective effect on one’s educational attainment and that of other family members (e.g., parents and offspring) (Greenfield & Moorman, 2019; Ross & Mirowsky, 2010; Wolfe et al., 2018; Yahirun et al., 2020). However, these findings are primarily based on U.S. data that excludes its territories and other studies focused on low- and middle-income countries that provide mixed results on the association between family education and health (Smith-Greenaway et al., 2018; Yahirun et al., 2017; Yang et al., 2016; Zimmer et al., 2002, 2007). Thus, the goal of the present study was to understand the role of multigenerational educational attainment and health among the older Puerto Rican population who has experienced transformative social and economic changes that have invariably shaped material resources across the life course. Drawing data from PREHCO, our results demonstrated a protective effect of maternal and paternal education on women’s health at baseline. However, this association was not significant when adding their own and their children’s education. Conversely, none of the three generations’ educational attainment was associated with chronic disease for men. However, experiencing economic deprivation as a child increased the average number of chronic conditions reported in older adulthood for men after accounting for all three generations’ education.

Prior research supports the associations between educational attainment and health through multiple socioeconomic and psychosocial pathways (Hummer & Lariscy, 2011; Jackson, 2009; Mirowsky & Ross, 2003). This line of inquiry also depicts a spillover effect where household members benefit from others’ educational attainment by granting access to instrumental resources and health-promoting knowledge that could potentially influence their living conditions and agency (Friedman et al., 2015; Greenfield & Moorman, 2019). For example, parents that have socioeconomic resources are able to increase the investments they can make for their children, which promotes a lower exposure to psychosocial hazards and decreases the development of adverse health conditions in later life (Erola et al., 2016; Goltermann et al., 2020; Khan et al. 2015). There is also an upstream benefit of children’s educational attainment where parents benefit from their children’s knowledge about health behaviors and health care navigating skills (Wolfe et al., 2018).

Our findings partially support previous results demonstrating a spillover effect of parental educational attainment among Puerto Rican women. This finding could be explained by the strong role gender norms play in the childhood socialization process (Cauce & Domenech-Rodríguez, 2002; Fuligni et al., 1999; Raffaelli & Ontai, 2004). Similar to other Latino cultures across the Americas, 20^th century Puerto Rican parents raised their children to follow distinctive conduct codes that set clear expectations of what activities and behaviors young women and men should embrace (Torres, 1998). For example, young women were expected to participate in domestic activities (e.g., cleaning, cooking, childcare) and stay indoors to limit their interactions with people outside the nuclear family circle. In contrast, young men were expected to join the labor force early to financially support their families, especially during economically constrained periods on the island.

For women, paternal and maternal education was associated with a decreased average number of chronic conditions, and this effect disappeared when we accounted for respondents and their children’s education. This loss of significance of parental schooling could exhibit a non-Markovian process, meaning there is a direct effect on a first generation’s socioeconomic attainment over both second and third generations’ socioeconomic mobility (Mare, 2011, 2014), and health (Huang et al., 2015; Lê-Scherban et al., 2014). Because of young women’s greater sensitivity to household socioeconomic context and family members’ influential role in their socialization (Moen, 2001; Morton & Ferraro, 2020), they could have benefited more from instrumental (e.g., economic stability) and psychosocial (e.g., emotional support) resources enabled by their parents’ education; thus, facilitating their socioeconomic attainment and later-life health (Mirowsky & Ross, 1998, 2003; Ross & Mirowsky, 2010). The results of our study may also suggest weaker educational gradients across adjacent generations due to the economic context of the island, which may result in the dilution of economic resources in families despite higher educational attainment in more recent generations (EDU20c, 2017; Rivera-Batiz & Santiago, 1996).

According to the gendered life course perspective, gender norms model different life trajectories for women and men that trigger significant contextual factors associated with later-life health outcomes (Moen, 2001). In the case of Puerto Rican women in the PREHCO sample, their parents’ education might have protected them from household stressors associated with cardiovascular diseases (McEwen, 1998; Pearlin et al., 1981; Wade et al., 2016) and encouraged their upward social mobility (Chevalier et al., 2013). Highly educated parents are more likely to embrace gender egalitarianism and support their daughters’ upward social mobility (Goldscheider et al., 2015; McDonald, 2000). Moreover, prior scholarship has shown greater health investment behaviors, including medical check-ups, healthy eating, and physical activity, among children of highly educated mothers (Prickett & Augustine, 2016).

These findings support a growing scholarship investigating how family educational resources have different individual health implications among low- and middle-income countries from what has been observed in the United States (Smith-Greenaway et al., 2018; Yang et al., 2016; Zimmer et al., 2002, 2007). For example, drawing data from the Mexican Health and Aging Study, Yahirun and colleagues (2017) tested the short- and long-term benefits of adult children’s education on older parents’ longevity in Mexico. They found greater sensitivity in mothers’ longevity. A more recent study also found greater benefits from adult children’s education on older Mexican mothers’ health cognition (Ma et al., 2021). As our research, their findings show how household educational resources are more influential on women’s health.

For Puerto Rican men in our sample, the detrimental effect of being economically deprived as a child might be due to their exposure to work hazards since they were expected to join the workforce to financially support their families alongside the head of household (Ibrahim et al., 2019; Wheleham, 1995). Most Puerto Rican men who joined the labor force as a child took jobs on plantation sites and endured unbearable heat and sun exposure and stressful confrontations with other workers; thus, increasing their chances of developing chronic conditions in adulthood (McEwen, 1998; Wright & Norval, 2021). Furthermore, these jobs were based on strenuous physical labor, such as carrying heavy objects without a reasonable resting time, that might have influenced the development of cardiometabolic problems in later life (Coenen et al., 2018).

We also find that both men and women born between 1916 and 1943 in our sample had an increased average number of chronic conditions compared to older birth cohorts (i.e., individuals born before 1916). A cohort selection effect may partially explain this since those from older cohorts may have better coped with the disadvantages of their generation due to their greater resiliency and robust physical and psychological well-being (Vaupel & Yashin, 1985a, 1985b). However, the results from the four-year follow-up among surviving female respondents showed no relationship between cohort status and the average number of chronic conditions reported. This may reflect the lack of cohort-based variation in chronic conditions among surviving female respondents or that these surviving respondents had greater opportunities and resources that delayed or lessened the impact of having a diagnosed health condition. In contrast, surviving male respondents in the youngest birth cohort (born from 1930 to 1943) had an increased average number of chronic conditions relative to the oldest birth cohort. Considering that surviving males who had experienced economic deprivation in childhood reported a lower average number of chronic conditions, it is possible that this form of childhood disadvantage converges with age due to selective mortality (House et al., 2005).

We also want to consider the possibility of the underreporting of chronic conditions that result from lower healthcare utilization among those who grew up with limited healthcare services (Mackenbach et al., 1996). Older cohorts of Puerto Ricans grew up in a period where healthcare centers were mainly located near the capital (i.e., San Juan) and trips to these establishments were long and dangerous because of limited road access, especially in rural areas (Mulligan, 2014). Thus, adults in these older birth cohorts learned to seek healthcare when strictly necessary (Yingwattanakul & Moschis, 2017). Another explanation may also include that the younger cohort of older adults in Puerto Rico is experiencing increases in life expectancy, contributing to increases in the incidence of disease (Crimmins et al., 2019).

Beyond the role of gender norms in establishing early life exposure to health-promoting and risky factors throughout the life course, Puerto Rico was subjected to a colonial context during the first half of the 20th century (Ayala & Bernabe, 2009). An agrarian economy dominated by a sugar oligopoly put additional financial constraints across households, limiting young Puerto Ricans’ educational opportunities (Dietz, 1986). Furthermore, each year, at the end of harvest (Zafra), rural families were heavily affected by the dead season (tiempo muerto), exposing them to economic deprivation and increasing class inequality (Ayala & Bergad, 2020). We can infer that these conditions exacerbated young Puerto Ricans’ health through household socioeconomic disparities that limited their access to health-promoting resources and mobility opportunities associated with their adulthood health (Elder, 1998; Phelan & Link, 2013). Moreover, our study reflects the challenges of using education as a socioeconomic measure in the context of Puerto Rico and other countries with comparable social and economic circumstances (e.g., Mexico, China, Taiwan) (Ma et al., 2021; Smith-Greenaway et al., 2018; Yahirun et al., 2017; Yang et al., 2016; Zimmer et al., 2002, 2007) where educational gradients are less pronounced and other socioeconomic individual factors and historical circumstances influence the relationship between education and health (Shavers, 2007). This is particularly relevant given that our longitudinal findings found no association between multigenerational educational attainment and health. This means that we need a better theoretical understanding of the social stratification of aging and health across the adult life in Puerto Rico to reduce socioeconomic disparities in health, especially earlier in the life course.

Limitations

While our study examined how multigenerational educational attainment influenced the health of a cohort of older Puerto Ricans with transformative educational changes throughout the 20^th century, the findings must be viewed considering several limitations. First, health data incorporated in our analysis were collected from self-reported doctor-diagnosed conditions, which assumes that an individual has interacted with the health care system for a doctor or physician to diagnose them with a health condition. Although most of the PREHCO respondents are insured (Pérez & Ailshire, 2017), this does not necessarily mean that there are increased interactions with the health care system, especially given the context of available physicians on the main island in areas outside of San Juan, Ponce, and Mayagüez. However, we want to note that we included self-reported medication use for controlling hypertension and diabetes, which increases the validity of these self-reported health conditions. Second, education may not be a good determinant of health in Puerto Rico in the way that it has been documented in the contiguous U.S. Puerto Rico has experienced multiple economic transitions and educational reforms that affected the instrumental and social meaning of this human capital measure (López Yustos, 1991; Rivera-Batiz & Santiago, 1996; Shavers, 2007). As a result, no recognized standards help define educational categories across generations for meaningful comparative purposes in Puerto Rico. A third limitation is the possibility of survivor bias. The respondents in PREHCO had to survive to older ages (minimum age 60) to be included in the study, which may be underestimating the associations between multigenerational educational attainment and health we document at baseline. Moreover, individuals included in the follow-up study also represent a selective survival of the sample, which may partially explain the null findings for parental education reported among females. Finally, the PREHCO study recruited community-dwelling older adults and excluded those living in nursing homes. Additionally, we excluded individuals with proxy interviews from our analytical sample. Thus, our sample may represent a relatively healthier proportion of the older Puerto Rican population.

Despite the study’s limitations, the topic of multigenerational effects of education and older adult health needs further exploration in the Latin American and Hispanic-Caribbean regions. Understanding the role of multigenerational educational changes of the 20^th century in Puerto Rico provides insight into past educational reforms and how drastic structural changes in socioeconomic Puerto Rican strata potentially generated long-term consequences on health, which is concerning given recent troubling trends in childhood education in Puerto Rico (see Appendix D for more information). Pursuing other outcomes, such as current mortality data in the PREHCO study when they become publicly available, and other health outcomes, will provide a more complete picture of the effects of multigenerational education on older adult health.

Footnotes

Spearman Correlation Matrix of Multigenerational Educational Attainment of Puerto Ricans Aged 60+ by Sex ( n =3238).

	Female respondents				Male respondents
	Father’s educational attainment	Mother’s educational attainment	Respondent’s educational attainment	Children’s educational attainment	Father’s educational attainment	Mother’s Educational Attainment	Respondent’s educational attainment	Children’s educational attainment
Father’s educational attainment	1.0000				1.0000
Mother’s educational attainment	0.5670***	1.0000			0.5462***	1.0000
Respondent’s educational attainment	0.4640***	0.4706***	1.0000		0.3744***	0.4383***	1.0000
Children’s educational attainment	0.2767***	0.3021***	0.3930***	1.0000	0.1786***	0.2518***	0.3906***	1.0000

Note. *p < .05; **p < .01; ***p < .001.

Rate Ratios (RR) With 95% Confidence Intervals (95% CI) for Chronic Conditions Among Surviving PREHCO Females,2006–2007 ( n =1470).

	Model 1			Model 2			Model 3			Model 4			Model 5
	RR		95% CI	RR		95% CI	RR		95% CI	RR		95% CI	RR		95% CI
Multigenerational educational attainment
Father’s education (ref = no education)
<8th grade	0.97		[0.84, 1.13]										0.94		[0.80, 1.11]
8th Grade	0.95		[0.79, 1.14]										0.94		[0.77, 1.15]
Mother’s education (ref = no education)
<8th grade				1.11		[0.97, 1.27]							1.11		[0.97, 1.27]
8th Grade				0.97		[0.81, 1.16]							0.97		[0.78, 1.22]
Respondent education (ref = 0–5 years)
6–11 years							1.02		[0.88, 1.18]				1.03		[0.88, 1.21]
12 or more years							1.03		[0.88, 1.20]				1.06		[0.89, 1.28]
Child’s education (ref = 0–12 years)
13–16 years										0.96		[0.85, 1.10]	0.95		[0.84, 1.08]
17+ years										0.94		[0.83, 1.07]	0.93		[0.81, 1.07]
Respondent childhood experiences
Economic deprivation							0.96		[0.87, 1.07]				0.95		[0.86, 1.05]
Left school to work							0.96		[0.82, 1.12]				0.97		[0.85, 1.12]
Demographics
Age	0.99	*	[0.98, 1.00]	0.99		[0.98, 1.00]	0.99		[0.98, 1.00]	0.99	*	[0.98, 1.00]	0.99		[0.98, 1.00]
Birth cohort (ref = 1901–1915)
1916–1929	1.13		[0.76, 1.66]	1.17		[0.79, 1.72]	1.14		[0.78, 1.66]	1.17		[0.79, 1.73]	1.18		[0.81, 1.73]
1930–1943	1.01		[0.65, 1.58]	1.04		[0.66, 1.63]	1.03		[0.66, 1.61]	1.05		[0.68, 1.63]	1.05		[0.68, 1.62]
Lived in the City/Suburbs/Elsewhere before age 18							0.95		[0.83, 1.08]				0.96		[0.84, 1.09]
Number of siblings							1.00		[0.98, 1.01]	1.00		[0.98, 1.02]	1.00		[0.98, 1.01]
Number of children										0.89	*	[0.80, 0.98]	1.00		[0.98, 1.02]
Sex of highest educated child is female													0.90	*	[0.82, 0.99]
Number of chronic conditions in wave 1	1.41	***	[1.35, 1.48]	1.42	***	[1.35, 1.49]	1.42	***	[1.36, 1.48]	1.41	***	[1.35, 1.48]	1.42	***	[1.36, 1.48]

Note. All variables are weighted using PREHCO sample weights; RR = Rate ratio; CI = Confidence interval; * p < .05; ** p < .01; *** p < .001.

Rate Ratios (RR) With 95% Confidence Intervals (95% CI) for Chronic Conditions Among Surviving PREHCO Males,2006–2007 ( n =865).

	Model 1			Model 2			Model 3			Model 4			Model 5
	RR		95% CI	RR		95% CI	RR		95% CI	RR		95% CI	RR		95% CI
Multigenerational educational attainment
Father’s education (ref = no education)
<8th grade	0.91		[0.75, 1.10]										0.92		[0.77, 1.10]
8th Grade	0.95		[0.76, 1.18]										0.96		[0.75, 1.23]
Mother’s education (ref = no education)
<8th grade				0.92		[0.76, 1.11]							0.91		[0.77, 1.09]
8th Grade				0.92		[0.74, 1.16]							0.90		[0.68, 1.18]
Respondent education (ref = 0–5 years)
6–11 years							1.18		[0.94, 1.47]				1.19		[0.95, 1.48]
12 or more years							1.01		[0.81, 1.26]				1.03		[0.82, 1.29]
Child’s education (ref = 0–12 years)
13–16 years										1.06		[0.92, 1.23]	1.08		[0.93, 1.25]
17+ years										1.00		[0.83, 1.21]	1.04		[0.86, 1.26]
Respondent childhood experiences
Economic deprivation							0.80	*	[0.67, 0.96]				0.80	*	[0.67, 0.95]
Left school to work							0.99		[0.81, 1.21]				0.99		[0.82, 1.20]
Demographics
Age	1.01		[0.99, 1.03]	1.01		[0.99, 1.03]	1.01		[0.99, 1.04]	1.01		[0.99, 1.04]	1.01		[0.99, 1.04]
Birth cohort (ref = 1901–1915)
1916–1929	1.63		[0.92, 2.88]	1.66		[0.94, 2.92]	1.68		[0.97, 2.90]	1.62		[0.91, 2.89]	1.69		[0.96, 2.98]
1930–1943	1.93		[0.94, 3.95]	1.97		[0.97, 4.02]	2.00	*	[1.02, 3.93]	1.90		[0.92, 3.94]	2.01	*	[1.01, 3.97]
Lived in the City/Suburbs/Elsewhere before age 18							0.93		[0.78, 1.11]				0.95		[0.79, 1.13]
Number of siblings							1.00		[0.98, 1.01]				1.00		[0.98, 1.01]
Number of children										1.00		[0.96, 1.03]	1.00		[0.97, 1.03]
Sex of highest educated child is female										0.99		[0.89, 1.11]	0.99		[0.90, 1.10]
Number of chronic conditions in wave 1	1.48	***	[1.37, 1.59]	1.47	***	[1.36, 1.59]	1.50	***	[1.39, 1.61]	1.47	***	[1.37, 1.58]	1.49	***	[1.39, 1.60]

Note. All variables are weighted using PREHCO sample weights; RR = Rate ratio; CI = Confidence interval; * p < .05; ** p < .01; *** p <.001.

Appendix D

Although this study examined a distinct cohort of older Puerto Ricans, it is alarming that the current school closures on the archipelago (that were exacerbated by Hurricane María) will influence access to and quality of education (Hinojasa et al., 2019), which will have reverberating impacts on life course chances, and in turn, long-term consequences on health for younger cohorts of Puerto Ricans that will eventually age. Already facing the biggest economic crisis and the most dangerous hurricane in the last one hundred years, Puerto Rico’s Department of Education closed 173 schools (Hinojasa et al., 2019). Overall, this measure has increased overcrowding, especially among elementary schools, potentially affecting teachers’ ability to provide quality education and their pupils’ academic performance. Within this large amount of schools, 65% of them formerly welcomed children from low-income rural communities (USCB, 2014) where public transportation services to urban areas are limited or unavailable. Given that currently open facilities are not located within walking distance from communities that used to benefit from a nearby school, students may not report to their classes on a regular basis (Allen et al., 2018). As a long-term effect, these changes could impact current students’ academic achievement (Shaaban & Reda, 2021), odds of attending college (Pike et al., 2014), and subsequently later-life health. Therefore, education policies should aim for immediate infrastructural changes to mitigate economic challenges, long-term investments to boost educational achievement and human capital development, and reduce exposure to psychosocial hazards (e.g., stress, family conflict) triggered by drastic changes in the school system.

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the National Institute on Aging of the National Institutes of Health (Grant #: 3R01AG064769-03S1).

ORCID iD

Alejandra Colón-López

References

Adler

N. E.

Newman

(2002). Socioeconomic disparities in health: Pathways and policies. Health Affairs, 21(2), 60–76. https://doi.org/10.1377/hlthaff.21.2.60

Allen

C. W.

Diamond-Myrsten

Rollins

L. K.

(2018). School absenteeism in children and adolescents. American Family Physician, 98(12), 738–744.

Assari

Chalian

Bazargan

(2020). Race, ethnicity, socioeconomic status, and chronic lung disease in the U.S. Research in Health Science, 5(1), p48. https://doi.org/10.22158/rhs.v5n1p48

Ayala

C. J.

Bergad

L. W.

(2020). Agrarian Puerto Rico: Reconsidering rural economy and society, 1899–1940. Cambridge University Press.

Ayala

C. J.

Bernabe

(2009). Puerto Rico in the American century: A history since 1898. The University of North Carolina Press.

Bauman

Graft

N. T.

(2003). Education attainment, 2000.

Bobonis

G. J.

Toro

H. J.

(2007). Modern colonization and its consequences: The effects of U.S. educational policy on Puerto Rico’s educational stratification, 1899-1910. Caribbean Studies, 35(2), 30–76. https://www-jstor-org.web.bisu.edu.cn/stable/25613115

Brunello

Fort

Schneeweis

Winter-Ebmer

(2016). The causal effect of education on health: What is the role of health behaviors? Health Economics, 25(3), 314–336. https://doi.org/10.1002/hec.3141

Cauce

A. M.

Domenech-Rodríguez

(2002). Latino families: Myths and realities. In Contreras

Kems

Neal-Barnett

(Eds.), Latino children and families in the United States: Current research and future directions (p. 25). Praeger Publishers.

10.

Cerin

Leslie

(2008). How socio-economic status contributes to participation in leisure-time physical activity. Social Science & Medicine, 66(12), 2596–2609. https://doi.org/10.1016/j.socscimed.2008.02.012

11.

Chevalier

Harmon

O’ Sullivan

Walker

(2013). The impact of parental income and education on the schooling of their children. IZA Journal of Labor Economics, 2(1), 1–22. https://doi.org/10.1186/2193-8997-2-8

12.

Cockerham

(2006). Society of risk-takers: Living life on the edge. Worth.

13.

Cockerham

(2013). Bourdieu and an update of health lifestyle theory. In Cockerham

W. C.

(Ed.), Medical sociology on the move: New directions in theory (pp. 127–154). Springer.

14.

Cockerham

(2017). Medical sociology (14th ed.). Routledge.

15.

Coenen

Huysmans

M. A.

Holtermann

Krause

Van Mechelen

Straker

L. M.

Van Der Beek

A. J.

(2018). Do highly physically active workers die early? A systematic review with meta-analysis of data from 193 696 participants. British Journal of Sports Medicine, 52(20), 1320–1326. https://doi.org/10.1136/bjsports-2017-098540

16.

Coleman

(2001). Population ageing: an unavoidable future. Social Biology and Human Affairs, 2001(7), 1–10.

17.

Consul

P. C.

Famoye

(1992). Generalized poisson regression model. Communications in Statistics - Theory and Methods, 21(1), 89–109. https://doi.org/10.1080/03610929208830766

18.

Courtenay

W. H.

(2000). Constructions of masulinity and their influence on men’s well-being: a theory of gender and health. Social Science & Medicine, 50(10), 1385–1401. https://doi.or/10.1016/S0277-9536(99)00390-1 http://www.postpartummen.com/pdfs/SS&M.PDF

19.

Creatore

M. I.

Glazier

R. H.

Moineddin

Fazli

G. S.

Johns

Gozdyra

Matheson

F. I.

Kaufman-Shriqui

Rosella

L. C.

Manuel

D. G.

Booth

G. L.

(2016). Association of neighborhood walkability with change in overweight, obesity, and diabetes. JAMA, 315(20), 2211. https://doi.org/10.1001/jama.2016.5898

20.

Crimmins

E. M.

Zhang

Y. S.

Kim

J. K.

Levine

M. E.

(2019). Changing disease prevalence, incidence, and mortality among older cohorts: The health and retirement study. The Journals of Gerontology: Series A, 74(Suppl 1), S21–S26. https://doi.org/10.1093/gerona/glz075

21.

Del Moral

(2018). The politics of education in Puerto Rico, 1898 – 1930 Colonial Lessons. The American Historian, May.

22.

Dietz

J. L.

(1986). Economic history of Puerto Rico: Institutional changes and capitalist development. Princeton University Press.

23.

Dye

B. A.

Mitchel

J. T.

(2010). Data management in oral health research. In Giannobile

W. V

Burt

B. A.

Genco

R. J.

(Eds.), Clinical research in oral health (pp. 103–122). Wiley-Blackwell.

24.

EDU20c . (2017). Puerto Rico: Educational attainment. http://edu20c.org/puerto-rico/

25.

ELA Puerto Rico . (1990). Ley Num. 68: Ley Orgánica del Departamento de Educación, aprobada el 28 de agosto de 1990. Departamento de Estado de Puerto Rico.

26.

Elder

G. H.

(1998). The life course as developmental theory. Child Development, 69(1), 1–12. https://doi.org/10.1111/j.1467-8624.1998.tb06128.x

27.

Elder

G. H.

Johnson

M. K.

Crosnoe

(2003). The emergence and development of life course theroy. In J. T. Mortimer, & Shanahan

M. J.

(Eds.), Handbook of the life course (pp. 3–19). Springer.

28.

Elliott

(2013). Gender differences in the determinants of distress, alcohol misuse, and related psychiatric disorders. Society and Mental Health, 3(2), 96–113. https://doi.org/10.1177/2156869312474828

29.

Erola

Jalonen

Lehti

(2016). Parental education, class and income over early life course and children’s achievement. Research in Social Stratification and Mobility, 44, 33–43. https://doi.org/10.1016/j.rssm.2016.01.003

30.

Esterrich

(2018). Concret and countryside: The urban and the rural in 1950s Puerto Rico. University of Pittsburgh Press.

31.

Famoye

(2015). A multivariate generalized poisson regression model. Communications in Statistics - Theory and Methods, 44(3), 497–511. https://doi.org/10.1080/03610926.2012.743565

32.

Flores

Krogstad

J. M.

(2019). Puerto Rico’s population declined sharply after hurricanes Maria and Irma. Pew Research Center.

33.

Friedman

E. M.

Mare

R. D.

(2014). The schooling of offspring and the survival of parents. Demography, 51(4), 1271–1293. https://doi.org/10.1007/s13524-014-0303-z

34.

Friedman

E. M.

Montez

J. K.

Sheehan

C. M. D.

Guenewald

T. L.

Seeman

T. E.

(2015). Childhood adversities and adult cardiometabolic health: Does the quantity, timing, and type of adversity matter? Journal of Aging and Health, 27(8), 1311–1338. https://doi.org/10.1177/0898264315580122

35.

Fuligni

A. J.

Tseng

Lam

(1999). Attitudes toward family obligations among American adolescents with Asian, Latin American, and European backgrounds. Child Development, 70(4), 1030–1044. https://doi.org/10.1111/1467-8624.00075

36.

Goldscheider

Bernhardt

Lappegard

(2015). The gender revolution: A framework for understanding changing family and demographic behavior. Population and Development Review, 41(2), 207–239. https://doi.org/10.1111/j.1728-4457.2015.00045.x

37.

Goltermann

Redlich

Grotegerd

Dohm

Leehr

E. J.

Böhnlein

Förster

Meinert

Enneking

Richter

Repple

DeVillers

Kloecker

Jansen

Krug

Nenadic

Brosch

Dipl-Psych

T. M.

Stein

Dannlowski

(2020). Childhood maltreatment and cognitive functioning: the role of depression, parental education, and polygenic predisposition. Neuropsychopharmacology. https://doi.org/10.1038/s41386-020-00794-6

38.

Graham

J. W.

Cumsille

P. E.

Shevoch

A. E.

(2013). Methods for handeling missing data. In Schinka

J. A.

Velicer

W. F.

Weiner

I. B.

(Eds.), Handbook of psychology: Research methods in psychology (pp. 109–141). John Wiley & Son Inc.

39.

Graham

J. W.

Olchowski

A. E.

Gilreath

T. D.

(2007). How many imputations are really needed? some practical clarifications of multiple imputation theory. Prevention Science, 8(3), 206–213. https://doi.org/10.1007/s11121-007-0070-9

40.

Greenfield

E. A.

Moorman

S. M.

(2019). Childhood socioeconomic status and later life cognition: Evidence from the wisconsin Longitudinal study. Journal of Aging and Health, 31(9), 1589–1615. https://doi.org/10.1177/0898264318783489

41.

Harris

Yang

Hardin

J. W.

(2012). Modeling underdispersed count data with generalized poisson regression. The Stata Journal: Promoting Communications on Statistics and Stata, 12(4), 736–747. https://doi.org/10.1177/1536867X1201200412

42.

Harvie

Howell

Evans

(2015). Can diet and lifestyle prevent brest cancer: What is the evidence? American Society of Clinical Oncology Education Books, 35(1), e66–e73. https://doi.org/10.14694/EdBook_AM.2015.35.e66

43.

Heeringa

S. G.

West

B. T.

Berglund

P. A.

(2017). Applied survey data analysis (2nd ed.). Chapman & Hall/CRC Press.

44.

Hinojasa

Meléndez

Serevino Pietri

(2019). Population decline and school closure in Puerto Rico. Center for Puerto Rican Studies.

45.

Holtermann

Hansen

J. V.

Burr

Søgaard

Sjøgaard

(2012). The health paradox of occupational and leisure-time physical activity. British Journal of Sports Medicine, 46(4), 291–295. https://doi.org/10.1136/bjsm.2010.079582

46.

Holtermann

Krause

van der Beek

A. J.

Straker

(2018). The physical activity paradox: six reasons why occupational physical activity (OPA) does not confer the cardiovascular health benefits that leisure time physical activity does. British Journal of Sports Medicine, 52(3), 149–150. https://doi.org/10.1136/bjsports-2017-097965

47.

House

J. S.

Lantz

P. M.

Herd

(2005). Continuity and change in the social stratification of aging and health over the life course: Evidence from a nationally representative Longitudinal study from 1986 to 2001/2002 (Americans’ Changing Lives Study). The Journals of Gerontology: Series B, 60(Special_Issue_2), S15–S26. https://doi.org/10.1093/geronb/60.Special_Issue_2.S15

48.

Huang

J. Y.

Gavin

A. R.

Richardson

T. S.

Rowhani-Rahbar

Siscovick

D. S.

Enquobahrie

D. A.

(2015). Are early-life socioeconomic conditions directly related to birth outcomes? Grandmaternal education, grandchild birth weight, and associated bias analyses. American Journal of Epidemiology, 182(7), 568–578. https://doi.org/10.1093/aje/kwv148

49.

Hummer

R. A.

Lariscy

J. T.

(2011). Educational attainment and adult mortality. January. https://doi.org/10.1007/978-90-481-9996-9

50.

Ibrahim

Abdalla

S. M.

Jafer

Abdelgadir

de Vries

(2019). Child labor and health: A systematic literature review of the impacts of child labor on child’s health in low- and middle-income countries. Journal of Public Health, 41(1), 18–26. https://doi.org/10.1093/pubmed/fdy018

51.

Jackson

M. I.

(2009). Understanding links between adolescent health and educational attainment. Demography, 46 (4), 671–694.

52.

Jiang

Kaushal

(2020). How children’s education affects caregiving: Evidence from parent’s last years of life. Economics and Human Biology, 38, 100875. https://doi.org/10.1016/j.ehb.2020.100875

53.

Khan

R.,

Iqbal

N.,

& Tasneem

(2015). The influence of parental education level on secondary school students academic achievement in district Rajanpur. Journal of Education and Practice, 6(16), 76–79.

54.

Lê-Scherban

Diez Roux

A. V.

Morgenstern

(2014). Associations of grandparental schooling with adult grandchildren’s health status, smoking, and obesity. American Journal of Epidemiology, 180(5), 469–481. https://doi.org/10.1093/aje/kwu154

55.

López Yustos

(1991). Historia documental de la education en Puerto Rico (Segunda ed.). Publicaciones Puertorriqueñas.

56.

Yahirun

Saenz

Sheehan

(2021). Offspring educational attainment and older parents’ cognition in Mexico. Demography, 58(1), 75–109. https://doi.org/10.1215/00703370-8931725

57.

Mackenbach

J. P.

Looman

C. W.

Van der Meer

J. B.

(1996). Differences in the misreporting of chronic conditions, by level of education: The effect on inequalities in prevalence rates. American Journal of Public Health1, 86(5), 706–711.

58.

Mangiavacchi

Piccoli

(2018). Parental alcohol consumption and adult children’s educational attainment. Economics and Human Biology, 28, 132–145. https://doi.org/10.1016/j.ehb.2017.12.006

59.

Mare

R. D.

(2011). A multigenerational view of inequality. Demography, 48(1), 1–23. https://doi.org/10.1007/s13524-011-0014-7

60.

Mare

R. D.

(2014). Multigenerational aspects of social stratification: Issues for further research. Research in Social Stratification and Mobility, 35, 121–128. https://doi.org/10.1016/j.rssm.2014.01.004

61.

McDonald

(2000). Gender equity in theories of fertility transition. Population and Development Review, 26(3), 427–439. https://doi.org/10.1111/j.1728-4457.2000.00427.x

62.

McEniry

(2011). Infant mortality, season of birth and the health of older Puerto Rican adults. Social Science & Medicine, 72(6), 1004–1015. https://doi.org/10.1016/j.socscimed.2010.08.026

63.

McEniry

Palloni

(2010). Early life exposures and the occurrence and timing of heart disease among the older adult Puerto Rican population. Demography, 47(1), 23–43. https://doi.org/10.1353/dem.0.0093

64.

McEniry

Palloni

Dávila

A. L.

Gurucharri

A. G.

(2008). Early life exposure to poor nutrition and infectious diseases and its effects on the health of older Puerto Rican adults. Journals of Gerontology - Series B Psychological Sciences and Social Sciences, 63(6), 337–348. https://doi.org/10.1093/geronb/63.6.S337

65.

McEwen

(1998). Stress, adaptation, and disease: Allostasis and allostatic load. Annals of the New York Academy of Sciences, 840(1), 33–44. https://doi.org/10.1111/j.1749-6632.1998.tb09546.x

66.

Menvielle

Boshuizen

Kunst

A. E.

Dalton

S. O.

Vineis

Bergmann

M. M.

Hermann

Ferrari

Raaschou-Nielsen

Tjonneland

Kaaks

Linseisen

Kosti

Trichopoulou

Dilis

Palli

Krogh

Panico

Tumino

Bueno-de-Mesquita

H. B.

(2009). The role of smoking and diet in explaining educational inequalities in lung cancer incidence. JNCI Journal of the National Cancer Institute, 101(5), 321–330. https://doi.org/10.1093/jnci/djn513

67.

Mirowsky

Ross

C. E.

(1998). Education, personal control, lifestyles and health: a human capital hypothesis. Research on Aging, 20(4), 415–449.

68.

Mirowsky

Ross

C. E.

(2003). Education, social status, and health. Adline Transaction.

69.

Moen

(2001). Gender, age, and the life course. In Binstock

R. H.

George

L. K.

(Eds.), Handbook of aging and the social sciences (5th ed., pp. 179–196). Academic Press.

70.

Morton

P. M.

Ferraro

K. F.

(2020). Early social origins of biological risks for men and women in later life. Journal of Health and Social Behavior, 61(4), 503–522. https://doi.org/10.1177/0022146520966364

71.

Mróz

L. W.

Chapman

G. E.

Oliffe

J. L.

Bottorff

J. L.

(2011). Men, food, and prostate cancer: Gender influences on men’s diets. American Journal of Men’s Health, 5(2), 177–187. https://doi.org/10.1177/1557988310379152

72.

Mulligan

J. M.

(2014). Unmanageable care: An ethnography of health care privatization in Puerto Rico. NYU Press.

73.

Palloni

Dávila

A. L.

Sánchez-Ayendez

(2013). Puerto Rican elderly: Health conditions (PREHCO) Project, 2002-2003, 2006-2007.

74.

Pantojas-García

(1990). Development strategies as ideology: Puerto Rico’s export-led industrialization experience. Lynne Rienner Publisher.

75.

Pearlin

L. I.

Menaghan

E. G.

Lieberman

M. A.

Mullan

J. T.

(1981). The stress process. Journal of Health and Social Behavior, 22(4), 337–356.

76.

Pérez

Ailshire

J. A.

(2017). Aging in Puerto Rico: A comparison of health status among Island Puerto Rican and Mainland U.S. Older Adults. Journal of Aging and Health, 29(6), 1056–1078. https://doi.org/10.1177/0898264317714144

77.

Phelan

J. C.

Link

B. G.

(2013). Fundamental cause theory. In Cockerham

W. C.

(Ed.), Medical sociology on the move: New directions in theory (pp. 127–154). Springer.

78.

Pike

G. R.

Hansen

M. J.

Childress

J. E.

(2014). The influence of students’ pre-college characteristics, high school experiences, college expectations, and initial enrollment characteristics on degree attainment. Journal of College Student Retention: Research, Theory and Practice, 16(1), 1–23. https://doi.org/10.2190/CS.16.1.a

79.

Poulton

Caspi

Milne

B. J.

Thomson

W. M.

Taylor

Sears

M. R.

Moffitt

T. E.

(2002). Association between children’s experience of socioeconomic disadvantage and adult health: A life-course study. Lancet, 360(9346), 1640–1645. https://doi.org/10.1016/S0140-6736(02)11602-3

80.

Prickett

K. C.

Augustine

J. M.

(2016). Maternal education and investments in children’s health. Journal of Marriage and Family, 78(1), 7–25. https://doi.org/10.1111/jomf.12253

81.

Raffaelli

Ontai

L. L.

(2004). Gender socialization in Latino/a families: Results from two retrospective studies. Sex Roles, 50(5/6), 287–299. https://doi.org/10.1023/B:SERS.0000018886.58945.06

82.

Rieker

Bird

C. E.

Laug

M. E.

(2010). Understanding gender and health: Old patterns, new trends, and future directions. In Bird

C. E.

Conrad

Fremont

A. M.

Timmermans

(Eds.), Handbook of medical sociology (pp. 52–74). Vanderbilt University Press.

83.

Rivera-Batiz

Santiago

C. E.

(1996). Island paraxod: Puerto Rico in the 1990s. Russel Sage Foundation.

84.

Rivera Median

Ramírez

R. L.

(1985). Del cañaveral a la fábrica: Cambio social en Puerto Rico. Ediciones Huracán-Acamemia.

85.

Rodriguez-Ayuso

Pesante-Gonzalez

F. J.

Geerman

(2012). Puerto Rico community health assessment: Secondary data analysis. http://www.salud.gov.pr/Datos/EstadisticasVitales/EstudiodeNecesidades/PRCHA_Secondarydataanalysis_final_2012.pdf

86.

Ross

C. E.

Mirowsky

(2010). Why education is the key to socioeconomic differentials in health. In Bird

C. E.

Conrad

Fremont

A. M.

Timmermans

(Eds.), Handbook of medical sociology (6th ed., pp. 33–51). Vanderbilt University Press.

87.

Ross

C. E.

Mirowsky

(2011). The interaction of personal and parental education on health. Social Science & Medicine, 72(4), 591–599. https://doi.org/10.1016/j.socscimed.2010.11.028

88.

Ross

C. E.

C.-L.

(1996). Education, age, and the cumulative advantage in health. Journal of Health and Social Behavior, 37(1), 104–120. https://doi.org/10.2307/2137234

89.

Rothgerber

(2013). Real men don’t eat (vegetable) quiche: Masculinity and the justification of meat consumption. Psychology of Men & Masculinity, 14(4), 363–375. https://doi.org/10.1037/a0030379

90.

Schnittker

(2004). Education and the changing shape of the income gradient in health. Journal of Health and Social Behavior, 45(3), 286–305. https://doi.org/10.1177/002214650404500304

91.

Seeman

T. E.

(1983). Health behavior and personal autonomy : A Longitudinal study of the sense of control in illness. Journal of Health and Social Behavior, 24(2), 144–160. https://doi.org/10.2307/2136641

92.

Shaaban

Reda

(2021). Impact of college provided transportation on the absenteeism and academic performance of engineering students. Eurasia Journal of Mathematics, Science and Technology Education, 17(3), Article em1944. https://doi.org/10.29333/ejmste/9727

93.

Shavers

V. L.

(2007). Measurement of socioeconomic status in health disparities research. Journal of the National Medical Association, 99(9), 1013–1023. https://doi.org/10.13016/avw3-9cvx

94.

Smith-Greenaway

Brauner-Otto

Axinn

(2018). Offspring education and parental mortality: Evidence from South Asia. Social Science Research, 76, 157–168. https://doi.org/10.1016/j.ssresearch.2018.07.001

95.

StataCorp (2019). Stata statistical software (No. 16). StataCorp LLC.

96.

Syamlal

Mazurek

J. M.

Dube

S. R.

(2014). Gender differences in smoking among U.S. working adults. American Journal of Preventive Medicine, 47(4), 467–475. https://doi.org/10.1016/j.amepre.2014.06.013

97.

Telfair

Shelton

T. L.

(2012). Educational attainment as a social determinant of health. North Carolina Medical Journal, 73(5), 358–365. https://doi.org/10.18043/ncm.73.5.358

98.

Thomas

P. A.

(2016). The impact of relationship-specific support and strain on depressive symptoms across the life course. Journal of Aging and Health, 28(2), 363–382. https://doi.org/10.1177/0898264315591004

99.

Torres

(1998). Masculinity and gender roles among puerto rican men: Machismo on the U.S. Mainland. American Journal of Orthopsychiatry, 68(1), 16–26. https://doi.org/10.1037/h0080266

100.

Torssander

(2013). From child to parent? The significance of children’s education for their parents’ longevity. Demography, 50(2), 637–659. https://doi.org/10.1007/s13524-012-0155-3

101.

USCB . (2014). Puerto Rico community survey - U.S.census Bureau, 2012. https://www.census.gov/content/dam/Census/newsroom/releases/2014/cb14-17_prcs.pdf

102.

U.S. Census Bureau . (2019). Annual estimates of the resident popualtion for selected age groups by sex for the United States, states, counties, and Puerto Rico.

103.

U.S. Census Bureau . (2021). Quick facts Puerto Rico.

104.

Van Cauwenberg

Van Holle

De Bourdeaudhuij

Van Dyck

Deforche

(2016). Neighborhood walkability and health outcomes among older adults: The mediating role of physical activity. Health & Place, 37, 16–25. https://doi.org/10.1016/j.healthplace.2015.11.003

105.

Vaupel

J. W.

Yashin

A. I.

(1985b). The deviant dynamics of death in heterogeneous populations. Sociological Methodology, 15, 179–211. https://doi.org/10.2307/270850

106.

Vaupel

J. W.

Yashin

A. I.

(1985a). Heterogeneity’s ruses: Some surprising effects of selection on population dynamics. The American Satistician, 39(3), 176–185.

107.

Wade

Cronholm

P. F.

Fein

J. A.

Forke

C. M.

Davis

M. B.

Harkins-Schwarz

Pachter

L. M.

Bair-Merritt

M. H.

(2016). Household and community-level adverse childhood experiences and adult health outcomes in a diverse urban population. Child Abuse & Neglect, 52, 135–145. https://doi.org/10.1016/j.chiabu.2015.11.021

108.

Wheleham

(1995). Modern feminist thought: From the second wave to ’post-feminism. New York University Press.

109.

Wolfe

J. D.

Bauldry

Hardy

M. A.

Pavalko

E. K.

(2018). Multigenerational attainments, race, and mortality risk among silent generation Women. Journal of Health and Social Behavior, 59(3), 335–351. https://doi.org/10.1177/0022146518784596

110.

Wolfson

J. A.

Ramsing

Richardson

C. R.

Palmer

(2019). Barriers to healthy food access: Associations with household income and cooking behavior. Preventive Medicine Reports, 13, 298–305. https://doi.org/10.1016/j.pmedr.2019.01.023

111.

Wright

C. Y.

Norval

(2021). Health risks associated with excessive exposure to solar ultraviolet radiation among outdoor workers in South Africa: An overview. Frontiers in Public Health, 9, 678680. https://doi.org/10.3389/fpubh.2021.678680

112.

Yahirun

J. J.

Sheehan

C. M.

Hayward

M. D.

(2017). Adult children’s education and changes to parents’ physical health in Mexico. Social Science & Medicine, 181, 93–101. https://doi.org/10.1016/j.socscimed.2017.03.034

113.

Yahirun

J. J.

Vasireddy

Hayward

M. D.

(2020). The education of multiple family members and the life-course pathways to cognitive impairment. The Journals of Gerontology: Series B, 75(7), e113–e128. https://doi.org/10.1093/geronb/gbaa039

114.

Yang

Martikainen

Silventoinen

(2016). Effects of individual, spousal, and offspring socioeconomic status on mortality among elderly people in China. Journal of Epidemiology, 26(11), 602–609. https://doi.org/10.2188/jea.JE20150252

115.

Yingwattanakul

Moschis

G. P.

(2017). Life course perspectives on the onset and continuity of preventive healthcare behaviors. The Journal of Primary Prevention, 38(5), 537–550. https://doi.org/10.1007/s10935-017-0482-7

116.

Zimmer

Hanson

H. A.

Smith

K. R.

(2016). Offspring socioeconomic status and parent mortality within a historical population. Demography, 53(5), 1583–1603. https://doi.org/10.1007/s13524-016-0502-x

117.

Zimmer

Hermalin

A. I.

Lin

H.-S.

(2002). Whose education counts? The added impact of adult-child education on physical functioning of older Taiwanese. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 57(1), S23–S32. https://doi.org/10.1093/geronb/57.1.S23

118.

Zimmer

Martin

L. G.

Ofstedal

M. B.

Chuang

Y.-L.

(2007). Education of adult children and mortality of their elderly parents in Taiwan. Demography, 44(2), 289–305. https://doi.org/10.1353/dem.2007.0020

20th Century Puerto Rico and Later-Life Health: The Association Between Multigenerational Education and Chronic Conditions in Island-Dwelling Older Adults

Abstract

Objectives:

Methods:

Results:

Discussion:

Keywords

Introduction

Background

Puerto Rico in the 20th and 21st Century

Education and Linked Lives

The Gendered Life Course

Study Purpose

Methods

Data

Measures

Chronic Disease

Educational Attainment

Childhood Experiences

Sociodemographics

Analytic Strategy

Results

Baseline Sample Characteristics

Educational Attainment and Chronic Conditions at Baseline

Educational Attainment and Chronic Conditions at Follow-Up

Discussion

Limitations

Footnotes

Spearman Correlation Matrix of Multigenerational Educational Attainment of Puerto Ricans Aged 60+ by Sex ( n =3238).

Rate Ratios (RR) With 95% Confidence Intervals (95% CI) for Chronic Conditions Among Surviving PREHCO Females,2006–2007 ( n =1470).

Rate Ratios (RR) With 95% Confidence Intervals (95% CI) for Chronic Conditions Among Surviving PREHCO Males,2006–2007 ( n =865).

Appendix D

Declaration of Conflicting Interests

Funding

ORCID iD

References

Puerto Rico in the 20^th and 21^st Century