Skip to main content
NCBI home page
SSM - Population Health logoLink to SSM - Population Health
. 2023 Mar 22;22:101380. doi: 10.1016/j.ssmph.2023.101380

The interaction of socioeconomic stress and race on telomere length in children: A systematic review and meta-analysis

Mariza Francis d, Alyssa Lindrose a, Samantha O'Connell e, Renee I Tristano b, Cecile McGarvey d, Stacy Drury a,b,c,d,∗
PMCID: PMC10102414  PMID: 37065841

Abstract

Rationale

Proposed mechanisms relating early life exposures to poor health suggest that biologic indicators of risk are observable in childhood. Telomere length (TL) is a biomarker of aging, psychosocial stress, and a range of environmental exposures. In adults, exposure to early life adversity, including low socioeconomic status (SES), is predictive of shorter TL. However, results in pediatric populations have been mixed. Defining the true relation between TL and SES in childhood is expected to enhance the understanding of the biological pathways through which socioeconomic factors influence health across the life span.

Objective

The aim of this meta-analysis was to systematically review and quantitatively assess the published literature to better understand how SES, race, and TL are related in pediatric populations.

Methods

Studies in the United States in any pediatric population with any measure of SES were included and identified through the following electronic databases: PubMed, EMBASE, Web of Science, Medline, Socindex, CINAHL, and Psychinfo. Analysis utilized a multi-level random-effects meta-analysis accounting for multiple effect sizes within a study.

Results

Thirty-two studies were included with a total of 78 effect sizes that were categorized into income-based, education-based, and composite indicators. Only three studies directly tested the relation between SES and TL as the primary study aim. In the full model, there was a significant relation between SES and TL (r = 0.0220 p = 0.0286). Analysis by type of SES categorization identified a significant moderating effect of income on TL (r = 0.0480, 95% CI: 0.0155 to 0.0802, p = 0.0045) but no significant effect for education or composite SES.

Conclusions

There is an overall association between SES and TL that is predominately due to the association with income-based SES measures implicating income disparities as a key target for efforts to address health inequity across the life span. Identification of associations between family income and biological changes in children that predict life-span health risk provides key data to support public health policies addressing economic inequality in families and presents a unique opportunity to assess the effect of prevention efforts at the biologic level.

Keywords: Telomere length, Meta-analysis, Socioeconomic factors, Children, Race

Highlights

  • •

    This meta-analysis finds that income-based assessments of SES are associated with shorter TL.

  • •

    Consistent measures of indicators of SES is needed to understand how low SES negatively effects child health.

  • •

    Addressing health disparities may require specific attention to income inequality.

  • •

    The effects of low SES on health can be detected at the molecular level in childhood.

1. Background

Despite having one of the highest net national incomes, the United States has one of the highest income inequalities in the world (Global inequality, 2021; National income, 2021). Inequalities of income, along with other indicators of Socioeconomic status (SES) such as wealth, educational attainment, economic position within a community, and financial stability, likely contribute to the persistence of health disparities across the life span in the United States and globally (Braveman, 2006; Chetty et al., 2016; Worthy, Lavigne, & Romero, 2020). Low SES is associated with a range of negative aging-related health outcomes, including earlier onset of diabetes, cardiovascular disease, and obesity in adults (He et al., 2001; Lantz et al., 1998; Robbins et al., 2001). Studies also consistently report that children from lower SES families have poorer health outcomes, matching observations in adult populations (Beebe-Dimmer, 2004; Galobardes, 2004; Kuh, 2002). A systematic review reported that poorer childhood socioeconomic position was associated with higher risk of all-cause mortality independent of adult socioeconomic position, suggesting that childhood SES exposures are predominant and persistent drivers of lifetime health risk (Galobardes, 2004; Galobardes, Lynch, & Smith, 2007). Proposed mechanisms include diminished access to appropriate health care and exposure to various stress exposures acrossacorss multiple settings (Cohen et al., 2010). These negative experiences in childhood set the trajectory for poorer long-term aging-related mental and physical health outcomes.

Although differences in SES contribute to health disparities, it is also evidence that racial and ethnic inequity, often amalgamated with SES, further amplify health disparities. . Blacks in the United States continue to have lower rates of educational attainment, almost 30% lower average household, and more than double the percentage of individuals living at the poverty level compared to non-Hispanic whites (Black/African American, 2021). Racial differences are also apparent in wealth, defined as the ability to plan for future economic needs and have financial reserves, with Hispanics and Blacks reporting less than 10% of non-Hispanic whites’ wealth (Williams, Priest, & Anderson, 2016). Even within individuals of similar SES racial disparities in health persist. For example, within high SES groups, there remain racial differences in the risk of heart disease mortality, diabetes, coronary artery disease, and hypertension (Williams et al., 2016). A significant effect of race on self-reported ill health and bed-days remains even after adjusting for SES measured by education, income, and occupation (Williams et al., 1997). Racial health disparities, independent of indicators of SES, have been reported in pediatric populations as well. Higher parental educational attainment and family income have protective health effects in non-Hispanic white children, but these same effects are not seen in Black or Hispanic children (Assari, 2020; Assari et al., 2020; Merkin et al., 2009). These differences are hypothesized to result from alterations to underlying biologic processes, within and across generations, driven by exposure to racism, discrimination, and structural inequality experienced by minoritized and marginalized populations of all ages (Bailey et al., 2017). The established associations of low SES and Black or Hispanic racial/ethnic inequity with disproportionate negative health outcomes hints at common biological pathways through which these exposures influence health risk across the life course.

Several theories and concepts provide overlapping insight into the potential biological pathways by which early life exposures predict and/or influence health across the life span. The developmental origins and health and disease (DoHAD) theorizes that biological changes, particularly epigenetic factors, triggered by environmental and psychosocial exposures early in life, including in utero, are key drivers of health across the life span. (Bianco-Miotto et al., 2017; Gluckman, Hanson, & Buklijas, 2010). Other commonly discussed models postulate biological mediators between early life exposures and health risk including allostatic load, toxic stress, and the “weathering hypothesis.” Allostatic load and toxic stress both theorize that how an organism balances, or fails to balance, repeated activation of the different stress resppnse systems due to shfits and stressors in the environment can drive earlier onset and more severe ill-health (Juster, McEwen, & Lupien, 2010; McEwen & McEwen, 2017). The “weathering hypothesis” specifically suggests that racism and structural inequality, within and across generations, disproportionately challenge the physiologic reserve of Black individuals, particularly women, resulting in accelerated aging and earlier disease onset that is strikingly apparent in the persistent disparities in perinatal outcomes experienced by women of color (Geronimus, 1992; Schmeer & Tarrence, 2018). Rooted in each of these models is that the early life differential exposures of minoritized groups to economic, racial, or traumatic stressors trigger physiologic and molecular changes which are observable in childhood and tied to accelerated aging and the earlier onset of age disorders (Berenson et al., 1992; Geronimus, 1992; Schmeer & Tarrence, 2018; Skinner et al., 2011; Theall et al., 2012a).

Telomeres, the repeated sequence of deoxyribonucleic acids (DNA) at the end of chromosomes critical to chromosome stability and genomic integrity, are an established biomarker of aging, psychosocial stress, environmental exposures, and early life adversity (Barraclough et al., 2019; Baskind et al., 2021; Epel et al., 2004; Hanssen et al., 2017; Haycock et al., 2014; Jiang et al., 2019; Miri et al., 2019; Oliveira et al., 2016; Ridout et al., 2018; Schutte & Malouff, 2014; Willeit et al., 2014). Telomere length (TL) decreases as a function of normal cellular replication but is further effected by oxidative stress, DNA damage, inflammation, and ionizing radiation (Broer et al., 2013). The relation between SES and TL has been examined in studies across the life span with mixed results. There is weak meta-analytic evidence of an association between TL and education level in adults, but little to no association with social class, income, or employment status (Robertson et al., 2013). Results from the U.S. Health and Retirement study, reported a weak association between SES and TL in adulthood, however lifetime cumulative adversity, particularly during childhood, had a strong association with shorter TL (Puterman et al., 2016). In one of the largest studies to date from the UK biobank a small, but significant, effect of various indicators of SES on adult TL was reported (Bountziouka et al., 2022). Together these data support several hypotheses: (1) the association between TL and SES may only be detectable in substantially large studies in adults due to a small effect size; (2) specific developmental windows exist where SES has distinct relevance to TL; or (3) TL may be uniquely associated with certain indicators of SES due to their stronger linkage to other forms of adversity. These hypotheses are not mutually exclusive and data from existing pediatric populations appear to favor a this more complex perspective (Puterman et al., 2016). For example while several pediatric studies reported TL associations with parental education, income, and neighborhood level indicators of SES (Needham et al., 2012; Theall et al., 2012b, Theall et al., 2017 other studies have failed to replicate these associations (Asok et al., 2013; Baskind et al., 2021; BosquetEnlow et al., 2020; Drury et al., 2014a; James et al., 2017; Kjaer et al., 2018; Ridout et al., 2019), providing the opportunity to explore variability in vulnerability across development.

Beyond developmental and methodologic approaches, variability in how SES is characterized may further contribute to differences in results. While many validated indicators of SES exist, they can be based on different indices such as annual household income, wealth, adult education level, and eligibility for government assistance. Education, income, and occupational class have been found to differentially predict the same health outcomes and a mortality risk (Duncan et al., 2002; Geyer, 2006). As such indicators of SES may reflect different aspects of the negative consequences of low SES, e.g. exposure to environmental contaminants such as lead, or elevated risk of exposure to community and neighborhood violence, that are not interchangeable or equivalent in their associations with health (Braveman et al., 2005). For example, in a socioeconomically and ethnically representative sample of US adults, low SES measured by education was associated with shorter TL while household income had no association with TL (Needham et al., 2013). Even when studies utilize the same metric of SES, for example household income, different categorization of income levels and/or significant differences in cost of living where the study is conducted, may influence comparability across studies (Shavers, 2007). Accounting for these variables meta-analytically is expected to provide a clearer estimate of the relationship between SES and child TL.

One additional consideration for studies examining the link between TL and SES is race. There is evidence of both racial differences in TL and effects of racism specifically on TL (Chae et al., 2014, Chae et al., 2016; Diez Roux et al., 2009; Drury et al., 2015a; Gardner et al., 2014; Jones et al., 2017; Zhu et al., 2011). Black infants have been found to have had significantly longer TL from blood spots than non-Hispanic white infants (Drury et al., 2015b). Leukocyte TL was found to be longer in Black adolescents compared to non-Hispanic white adolescents (Zhu et al., 2011). In adults, some studies have reported longer TL in Black and Hispanic men and women compared to non-Hispanic white, while others have reported shorter TL in minoritized populations hinting that racial differences in TL may change across the life span (Brown, Needham, & Ailshire, 2017; Diez Roux et al., 2009). The findings in adults suggest either accelerated TL shortening in individuals with longer TL at birth and/or accelerated TL shortening in minoritized individuals (Brown et al., 2017; Diez Roux et al., 2009; Needham et al., 2013). Additional studies have linked exposure to racism as an independent factor predicting shorter TL in minoritized adults which may further contribute to racial differences in TL (Chae et al., 2014, Chae et al., 2016; Pantesco et al., 2018). To better understand how race, SES, and TL relate, this meta-analysis systematically reviewed and quantitatively assessed the published literature in pediatric populations. We hypothesized that race, type of SES indicator, and age would be moderators of the association between SES and TL.

2. Methods

2.1. Database search & terms

Eligible studies were identified through a systematic search in PubMed, EMBASE, Web of Science, Medline, Socindex, CINAHL, and Psychinfo. Extensive key words and MeSH terms for children, socioeconomic status, and TL were included in the database search (Kachmar et al., 2019; Kastner et al., 2006). All databases were searched from 1988 to April 19, 2021. The search was first completed in PubMed (Supplemental Methods) and translated to additional databases.

2.2. Inclusion criteria

Articles were included if they were published in a peer-reviewed journal, measured TL in a pediatric population (2 months < age≤18 years) within the US, and measured any indicator of socioeconomic status and provided the correlation between SES and TL. As the first method to measure human TL was published in 1988, only articles published in 1988 or later were included (Moyzis et al., 1988).

2.3. Exclusion criteria

All review articles, abstracts, dissertations, and case reports were excluded. Articles were excluded if there was no measure of TL, if TL was measured only in adults (>18 years) or newborns (<2 months), TL measured only in cultured cells or tumors, in a study population outside of the US, if there was no indicator of SES, or if TL was only measured in postmortem tissue. During title and abstract screening, articles were retained even if the abstract/title did not specify a TL or SES relationship. Articles that did not include both TL and SES measurement were subsequently excluded during the full-text screening. Articles were also excluded if TL was measured in mixed-age populations without a pediatric subgroup. We also excluded studies in children with cancer, genetic conditions, or telomeric disorders.

2.4. Study selection process

Titles and abstracts of articles were screened in duplicate for exclusion criteria by independent reviewers (MF, ARL, CBM, RIT, SSD) using Covidence (Covidence). Conflicts were resolved via discussion by three reviewers (MF, ARL, SD). Articles that met eligibility criteria following title and abstract screening were subsequently reviewed in full text by independent reviewers (MF, ARL, CBM, RIT) in duplicate to verify eligibility. Given that an article may be excluded for multiple reasons, a hierarchy of exclusion criteria was created as follows to identify the primary reason for exclusion: 1. Study population outside of the US, 2. TL in adult population (>18), 3. TL in newborns (<2 months) or cord blood or placenta, 4. No TL measurement, 5. No SES measurement, 6. Other reasons such as TL measured in mixed-age population without a pediatric subpopulation analysis, 7. TL measured only in cultured cells, cell lines, tumors, cancer patients, or patients with a known telomere-related genetic disorder (e.g. dyskeratosis congenita). Conflicts were resolved by discussion between two reviewers (MF and ARL).

2.5. Data extraction

Data extraction was completed in duplicate from included studies by independent reviewers (MF, ARL, CBM, RIT). Information about study design, demographics, SES measures, TL measurement, and the association between SES and TL was extracted.

2.6. Study quality and risk of bias

A study quality assessment adapted from the Joanna Briggs Institute (JBI) critical appraisal tools and TL measurement assay quality assessment adapted from the Telomere Research Network reporting guidelines was completed for each article (Lindrose & Drury, 2020; Moola et al., 2020). The study quality assessments included questions about whether the inclusion criteria for the sample was clearly stated (97%, n = 31), if the study subjects and setting were adequately described (100%, n = 32), and if details about samples used in longitudinal studies were provided (60%, n = 3) (Moola et al., 2020). The TL quality assessment included questions about reporting of sample type, storage methods, TL assay protocols, details about how TL was determined, and details about assay precision (Lindrose & Drury, 2020). If the study included each TL quality criteria, it was assigned one point, and the quality of TL assay scores ranged from 4 to 7.

2.7. Contacting authors and inclusion of multiple studies from single cohorts

Authors were contacted if the primary article did not provide the correlation coefficients or beta coefficients between SES measures and TL (Asok et al., 2013; Baskind, 2021; BosquetEnlow et al., 2020; Carroll et al., 2020; Chen et al., 2020; Dismukes et al., 2016; Drury et al., 2014a, Drury et al., 2014b; Entringer et al., 2015; Esteves et al., 2020; Etzel et al., 2020, p. 120; Guarneri‐White et al., 2018; HenjeBlom et al., 2015; Humphreys et al., 2020; Kjaer et al., 2018, Kjaer et al., 2020; Kliewer & Robins, 2021; Koss et al., 2020; Kroenke et al., 2011; Lee et al., 2019; Manczak & Gotlib, 2020; Merrill et al., 2017; Miller et al., 2020; Mitchell et al., 2014, 2017; Nelson et al., 2018, Nelson et al., 2020, Nelson et al., 2021; Ridout et al., 2019; Robles et al., 2016; Sosnowski et al., 2021; Theall et al., 2013, 2017, 2019; Wojcicki et al., 2016, Wojcicki et al., 2018; Zeiger et al., 2018; Zhu et al., 2011, Zhu et al., 2014). Authors were additionally asked to confirm the sample size and racial/ethnic breakdown of the sample used to determine the correlation if it was not specified in the article. If all necessary data was provided in the manuscript, authors were not contacted (James et al., 2017; Needham et al., 2012). There are five cohorts that were used in multiple papers. One cohort was used in three papers, and the authors provided the overall dataset which was used once in the analysis (Chen et al., 2020; Zhu et al., 2011, Zhu et al., 2014). The same sample and SES measure from one cohort was used in two papers, and therefore, this data was only included once in the analysis (Nelson et al., 2018, Nelson et al., 2020). For three cohorts, papers that only utilized a subset of the entire cohort or reported on different SES measures (n = 13) or utilized a different tissue type for TL measurement (n = 2) were analyzed as unique data points in the meta-analysis.

2.8. Data analysis

Statistical analysis was conducted using a multi-level random-effects meta-analysis using RStudio Version 1.4.1717 and the metafor package to account for multiple effect sizes within the same study and multiple studies within the same cohort (R: The R Project for, 2021; Viechtbauer, 2010). The correlation coefficients (r) were used to determine the overall pooled effect size of the SES and TL relationship. These correlations were z-transformed to approximate a normal distribution for analysis with the esc package and converted back to correlation coefficients for presentation (esc package, 2021). To account for directional differences in how SES was categorized within each included study, (i.e., higher vs lower SES categories being considered the reference), we multiplied the outcomes where a higher SES category was the reference by −1 to assign appropriate directionality. This approach allowed all outcomes to trend in the same direction-specifically as each indicator of SES reflected greater wealth/education/income, the number reflective of the SES metric increased. A three-level model was used to account for within-study sampling variance (level 1), variance between effect sizes within the same study (level 2), and variance between studies (level 3). As our analysis included multiple articles utilizing samples from the same cohorts, a four-level model incorporating variance between cohorts was tested.

Moderator analyses were conducted, based on our apriori hypotheses, within the three-level model to examine the possible effect of the following variables as moderators of the overall effect: age, sample size, commonly used SES measures (education, income, and composite SES), whether SES and TL were measured concurrently, tissue source for TL (buccal, saliva, leukocyte, dried blood spot (DBS)), and heterogeneous or homogeneous participant racial demographics. Continuous predictors assessed were mean age, proportion of female participants, and proportion non-white participants. Each variable was assessed for significance in a univariate model. The model was also run within a subgroup of studies based on any significant moderator. Additionally, the model was run within a subgroup of studies based on their racial make-up.

2.9. Test for heterogeneity

The heterogeneity among effect sizes was determined using Q and I2 statistics. The Q statistic provides a test indicating whether the observed range of effect sizes is larger than what is expected from within-study variance alone and provides an indicator about the significance of these differences in effect size (Huedo-Medina et al., 2006). The I2 index provides a percentage of total variance that can be attributed to between-study variance and indicates the amount of between-study heterogeneity. I2 values of 25%, 50%, and 75% correspond with low, moderate, and high heterogeneity between-studies (Huedo-Medina et al., 2006). Low I2 values suggest that the variability between effect sizes of different studies is mostly attributable to sampling error rather than true heterogeneity.

2.10. Test for publication bias

Publication bias was assessed using the Egger's test and visualized using a funnel plot. The Egger's test regresses the standardized effect sizes on their precision and provides a level of significance of publication bias (Lin & Chu, 2018). The funnel plot presents effect sizes plotted against their standard errors (Lin & Chu, 2018).

3. Results

3.1. Screening

The database search identified 6519 articles for abtract and title screening after removal of duplicates. There were 558 articles assessed for eligibility via full-text screening, and 519 were excluded, resulting in 41 articles (Fig. 1). Nine papers were excluded in the final analysis because either the necessary data could not be provided or they were excluded because they were a duplicate study from the same cohort, resulting in a final inclusion of 32 studies.

Fig. 1.

Fig. 1

Open in a new tab

Prisma flow diagram

3.2. Data extraction

The characteristics of the 32 included articles including the SES indices utilized, age range of the population, sample size, longitudinal or cross sectional study design, percent non-white race, tissue source for telomere measurement, TL assay method, and whether or not TL and SES were measured concurrently are presented in Table 1. The age range of the pediatric participants ranged from four months to 18 years. Sample size of included studies ranged from 39 to 2527 with twenty-six papers including a sample size of <300 and 6 papers having a sample size >300 and all studies utilized a PCR based measurement assay Only 4 studies were longitudinal and half of the studies did not measure TL and SES concurrently.

Table 1.

Study characteristics of the thirty two studies included in the meta-analysis.

Author and Year Study Design Sample Size Age Range % Non-White Race % Female Tissue TL Assay SES Measurement(s) SES and TL concurrent?
Asok 2013 CS 89 4–7 89.4 44.9 buccal qPCR household income Y
Baskind 2021 CS 73 3–5 100.0 55.0 leukocyte qPCR maternal education, household income N
Bosquet-Enlow 2020 L 630 0.4–3 28.2 47.8 saliva qPCR parental education, annual household income, financial reserves N
Carroll 2020 CS 111 3–5 71.1 55.0 buccal qPCR maternal education, household income N
Chen 2020 CS 660 14–18 44.7 51.0 leukocyte qPCR Hollingshead Index Y
Dismukes 2016 CS 102 5–15 100.0 55.5 buccal MMqPCR maternal education Y
Drury 2014 CS 77 5–15 100.0 49.4 buccal MMqPCR maternal education Y
Drury 2014 CS 80 5–15 100.0 49.0 buccal MMqPCR maternal education, monthly household income, parent marital status Y
Entringer 2015 CS 100 0–3.8 59.0 43.0 buccal qPCR maternal education N
Esteves 2020 L 155 0.3–1.5 63.3 46.5 buccal MMqPCR SES index: education, employment status, home ownership, income, savings, and receipt of government assistance N
Etzel 2020 CS 124 NR 67.5 54.8 buccal qPCR education, household income, Y
Guarneri-White 2018 CS 108 11–19 41.7 55.6 saliva aTL class based on parental occupation Y
James 2017 CS 1567 9 77.2 47.6 saliva aTL maternal education, household income N
Kjaer 2018 L 102 4–5 100.0 NR DBS qPCR maternal education N
Kliewer 2021 CS 108 13–17 100.0 57.4 blood qPCR maternal education, family income Y
Koss 2020 CS 1194 9 79.3 100.0 saliva qPCR maternal education, family income, mother married to father at birth N
Kroenke 2011 CS 78 NR 43.0 58.9 buccal qPCR maternal education N
Manczak 2020 CS 122 9–15 36.0 100.0 saliva qPCR maternal education, average household income Y
Merrill 2017 CS 77 5–15 100.0 53.8 buccal MMqPCR maternal education Y
Miller 2020 L 214 9–15 51.3–54.6 56.5 saliva qPCR annual family income N
Mitchell 2014 CS 40 9 100.0 0.0 saliva qPCR maternal education, average income/needs ratio, family structure N
Mitchell 2017 CS 2420 9 79.0 48.0 saliva aTL maternal education, household income, mother lives with child's father at birth N
Needham 2012 CS 70 7–13 50.0 52.0 whole blood qPCR maternal education, annual household income Y
Nelson 2020 CS 48 NR 20.9 60.4 saliva qPCR household income N
Nelson 2021 CS 180 11–14 21.2 46.7 saliva qPCR household income Y
Ridout 2019 L 256 3–5 60.4 52.3 saliva qPCR socioeocnomic adversity, qualified for public assistance N
Robles 2016 CS 39 8–13 64.5 59.6 leukocyte qPCR parental education, household income Y
Sosnwoski 2021 CS 2527 8–10 84.0 48.0 saliva aTL parental education, parental income, Survey of Income and Program Participation and Social Indicators Survey N
Theall 2013 CS 99 4–14 100.0 52.9 saliva qPCR maternal education, Hollingshead Index, neighborhood disorder Y
Theall 2017 CS 85 4–14 100.0 58.8 saliva MMqPCR maternal education Y
Theall 2019 CS 90 5–16 100.0 54.0 buccal MMqPCR maternal education Y
Wojcicki 2018 CS 61 2–3 100.0 51.7 leukocyte qPCR maternal education, household income N
Open in a new tab

Three papers directly tested the relationship between SES and TL as the primary study aim (Mitchell et al., 2014; Needham et al., 2012; Theall et al., 2013). The remaining studies included SES as a covariate. All papers assessed SES at the household level, except for one that utilized a neighborhood level measure (Theall et al., 2013). More than half (59.4%, n = 19) of the papers included multiple measures of SES. Parental education was the most common metric (75.0%, n = 24), followed by some measure of income (59.4%, n = 19). All studies that included parental education measured maternal education, with only three studies also including paternal education. Variable measures of income were utilized across studies including annual household income, monthly household income, maternal income, paternal income, and income/needs ratio. Some papers (12.5%, n = 4) utilized a composite SES index, with two studies using the Hollingshead Index (Chen et al., 2020; Theall et al., 2013) and two using a customized SES index (Esteves et al., 2020; Ridout et al., 2019). Other less frequently used metrics of SES included family structure (12.5%, n = 4), employment/occupation status (3.1%, n = 1), neighborhood disorder (3.1%, n = 1), poverty assessment (3.1%, n = 1), qualified for public assistance (3.1%, n = 1), financial reserves (3.1%, n = 1), and type of health insurance (3.1%, n = 1).

SES and TL were measured concurrently in half (50.0%, n = 16) of the studies. The range of time between SES measurement and TL measurement for the remaining studies was 4 months–9 years. Five studies (31.3%, n = 5) measured TL at several timepoints, with SES only measured once prior to TL (BosquetEnlow et al., 2020; Esteves et al., 2020; Kjaer et al., 2018; Miller et al., 2020; Ridout et al., 2019). Eleven studies (34.4%, n = 11) measured TL and SES at different time points, and in all cases TL was measured at a later time point than the assessment of SES (Baskind et al., 2021; Carroll et al., 2020; Entringer et al., 2015; James et al., 2017; Koss et al., 2020; Kroenke et al., 2011; Mitchell et al., 2014, Mitchell et al., 2017; Nelson et al., 2020; Sosnowski et al., 2021; Wojcicki et al., 2018).

Various tissue sources were used for measurement of TL: saliva (43.8%, n = 14), buccal (34.4%, n = 11), leukocyte DNA (12.5%, n = 4), whole blood (6.3%, n = 2), and dried blood spots (DBS) (3.1%, n = 1). All studies utilized a quantitative PCR-based method for TL measurement: singleplex quantitative PCR (qPCR) (65.6%, n = 21), monochrome multiplex qPCR (MMqPCR) (21.9%, n = 7), and absolute TL (aTL) (12.5%, n = 4).

The racial composition varied across studies. Twenty studies had heterogenous racial makeup (percent non-white ranging from 20.9% to 89.4%), nine studies had a 100% Black sample, and three studies had a 100% Hispanic sample.

3.3. Study quality and risk of bias

All included articles described study subjects and study setting in detail (Table 1). All but one article had clear inclusion criteria for their participants (Guarneri‐White et al., 2018). Scores for telomere assay reporting quality ranged from 29% to 100% of reporting criteria (Supplemental Table 1). All studies reported the sample type used for TL measurement. Most studies reported the TL assay protocol (94%, n = 30) and the method of TL assay used (88%, n = 28), however fewer studies reported DNA extraction, storage, and processing information (75%, n = 24), number of sample replicates (78%, n = 25), and metrics of assay precision and reliability (69%, n = 22). Only rarely did articles report on sample storage conditions prior to DNA extraction (34%, n = 11).

3.3.1. Meta-analysis

A total of 78 effect sizes were included in the analysis with 34 correlations between education and TL, 23 between income and TL, and 21 between other SES measures and TL (see Data Extraction results) had 21 correlation coefficients. As the four-level model did not significantly strengthen the overall model when compared to the three-level model by ANOVA, the three-level model was used for the analysis. Using the three-level model, the relationship between SES and TL was significant (r = 0.0220 p = 0.0286) (Fig. 2). the variance between studies (V = 0.0001) and the variance between effect sizes within a study (V = 0.0006) were small. While within-study variance was small, after conducting the likelihood ratio test, comparing the full model to a reduced model in which within-study variance was set to zero, the within-study variance was significant (p = 0.0485), concluding that the fit of the three level model is statistically better than the reduced model as there is significant variability within studies. Between-study variance tested with the likelihood ratio test was not significant (p = 0.7955).

Fig. 2.

Fig. 2

Open in a new tab

Forest plot depicting the relation between SES and TL across studies.

3.4. Heterogeneity

The results of the test of heterogeneity (Q = 50.5252, p = 0.9915) indicated that there was no significant heterogeneity between effect sizes in the data set. The distribution of the variance, I2 index, was estimated across the three levels: within-study sampling variance (level 1), between effect sizes within studies (level 2), and between studies (level 3). Level one accounted for 85.8% of the total variance, level two accounted for 12.4% of the total variance, and level three accounted for 1.8% of the total variance indicating that the majority of variance was the result of within-study sampling error.

3.5. Moderator analysis

Given previous findings of differential associations of distinct categories of SES with TL, analysis within each type of SES categorization (e.g. income, education, and composite) was conducted. Income had a significant moderating effect (r = 0.0486, 95% CI: 0.0158 to 0.0812, p = 0.0044). The full model was significant when run within only the studies (n = 17, 23 correlation coefficients) that measured income (r = 0.0504, 95% CI: 0.0207 to 0.0800; p = 0.0019) (Fig. 3). Neither education nor composite SES had a significant moderator effect. Analyses run within studies that only measured education or composite SES did not have a significant moderating effect (Supplemental Figs. 1 and 2).

Fig. 3.

Fig. 3

Open in a new tab

Forest plot depicting the relation between SES and TL across studies.

Subsequent post-hoc analyses were conducted within the studies with income to determine if there was any subgroup moderation. Visualization of effect sizes for studies measuring income specifically and TL are shown in Fig. 3. Income and TL that were not measured concurrently significantly deviated from zero (r = 0.0481, 95% CI: 0.0143 to 0.0816; p = 0.0074), however income and TL measured concurrently did not. Similarly, in studies with an older mean age (>6) results significantly deviated from zero (r = 0.0619, 95% CI: 0.0325 to 0.0911; p = 0.0003), while studies with a <6 mean age did not. Studies with a heterogenous racial makeup significantly deviated from zero (r = 0.0512, 95% CI: 0.0208 to 0.0814, p = 0.0022), however studies with a homogenous racial group (Black or Hispanic) did not. Sample size, tissue type, and gender did not have any significant effects on the income and TL relationship.

3.6. Publication bias

The Egger's test and funnel plot (Fig. 4) did not indicate the presence of asymmetry (z = 0.027, 95% CI: −3.94 to 3.99, p = 0.99), suggesting little evidence of publication bias.

Fig. 4.

Fig. 4

Open in a new tab

Funnel plot (socioeconomic status and telomere length).

4. Discussion

This meta-analysis evaluated the relation, in childhood, between SES and TL. Data was pooled from 32 published papers and included 78 relevant effect sizes. Consistent with our original hypothesis, we found a significant overall association between SES and TL. There was no indication of publication bias and no effect of study sample size, telomere measurement assay, source of DNA for TL measurement, child sex, or racial composition of the cohort. However, we did find substantial variation in the metrics utilized to quantify SES. Given the sginficant within-study heterogeneity and the variability in the characterization and measurement of SES across studies, we further evaluated the relation of TL with distinct categories of SES.

Based on empirical findings related to other health outcomes, we tested moderation by SES categorized as income, education, or composite indicators. Only SES characterized by income resulted in a significant effect with high income associated with longer TL. Post-hoc moderator analyses revealed that the association between income and TL was driven by studies in older children (>6), studies that measured income prior to assessment of TL, and in studies with a heterogenous racial mixture. All effect sizes were small. The relation in older children and in studies measuring TL later are consistent with data across other health outcomes suggesting that early life exposure to socioeconomic-related stressors is predictive of later health outcomes, some of which may not be initially detectable, and also consistent with studies in adults in which childhood SES predicted TL in adulthood (Beebe-Dimmer, 2004; Galobardes, 2004; Galobardes et al., 2007; Khan et al., 2022; Kuh, 2002). Our results are also consistent with recent data from the UK biobank study of over 422,000 individuals, predominately white, where various indicators of SES were associated with small, but significant, effects on TL (Bountziouka et al., 2022). The detrimental biological and health effects of early SES associated stress likely develop over time within an individual (Luby et al., 2020).

Education and composite SES were not significant moderators indicating a need for caution when drawing comparisons across studies that utilize differing measures of SES. These differences also provide preliminary evidence that the biological correlates of SES related strain may differ based on how wealth or position within a society is defined. Although the evidence of the biological embedding of low income in childhood is highly concerning, a more optimistic perspective is to view this window as an opportunity for approaches focused on stabilizing TL decline early in life via enhanced income support to families as well as interventions focused on increasing positive health behaviors (e.g. exercise and mindfulness) or increase social connectedness which have growing evidence of ties to TL. Interventions, particularly in high-risk children being raised in low SES households, that seek to decrease exposure to known predictors of accelerated TL shortening, such as trauma, smoking, environmental toxins, or target the early prevention of diabetes and obesity, which are also well-established correlates of shorter TL, may also provide crucial buffering of the exposure to early socioeconomic stress on TL.

Tissue source and type of PCR-based TL assay did not moderate the relation between SES and TL. The lack of moderation by TL assay is unsurprising as all methods were qPCR based and, as such, measure an estimate of mean TL and have similar methodologic strengths and weaknesses (Lindrose & Drury, 2020). Studies utilizing other TL assays, such as TRF or TeSLA, may detect larger effect sizes but, the potential increased ability to detect small effect sizes must be balanced with the practicality and costs of qPCR based methods which require far less DNA, are generally higher throughput, and are less expensive (Lin et al., 2019). The lack of moderation by tissue type, while not entirely unexpected, could be due to either a high cross-tissue correlation of TL, particularly early in development, and/or a broad effect of income differences on telomere dynamics across tissues (Kuh, 2003).

Despite our hypothesis, we did not find an effect of race, nor did race moderate the relation of overall SES and TL. However, post-hoc analyses in mixed racial groups revealed that race moderated the association between TL and income-based SES. Specifically, there was no effect of income on TL, while within non-Hispanic white children there was an association. This finding is consistent with the hypothesis that the drivers of health risk for minoritized youth have less to do with income and are more likely due to other stressors and exposures tied with generations of racial inequity and structural racism.

4.1. Limitations

While there are strengths to this meta-analysis, there are also limitations. First, this meta-analysis included studies in which SES was not the primary exposure variable of interest, but rather a covariate. Only three studies looked at SES as the primary predictive factor of child TL (Mitchell et al., 2014; Needham et al., 2012; Theall et al., 2013) and all three found a significant association between TL and SES measures. One (Needham et al., 2012) showed that participants from low-income households or whose parents never attended college had shorter TL, one (Mitchell et al., 2014) showed significant associations between low income, lower maternal education, unstable family structure, and harsh parenting and TL, and one (Theall et al., 2013) showed significant associations between neighborhood variables of high disorder and percent below poverty and shorter TL. The findings in all three of these studies of an association between TL and SES indexed by income is consistent with our findings in the sub-group analyses of income-based SES. Recognizing that studies specifically examining an alternative predictor of TL may induce an unidentified bias, it is plausible that with sufficient numbers of studies that directly test the relation between SES and TL an overall significant association would have been detected when including other indicators of SES. We also included only studies conducted within the United States. This decision arose due to considerations of the challenges of translating SES indicators across countries and cultures (David, 2002). As such, our findings related to income-based SES and TL is not generalizable to other countries. Future directions that include studies across cultures, as well as longitudinal studies looking at SES as the primary predictive factor of TL across childhood, are needed. Additionally, a more comprehensive understanding of the relationship between different SES measures would be useful in understanding which measures are most predictive of health and how these biological pathways may differ.

4.2. Conclusion

This is the first meta-analysis to quantify the association between SES and TL in children. Our findings indicate that income-based SES is associated with TL and may have unique relevance in older children and within specific racial/ethnic groups. This specificity of the association with income-based SES is particularly concerning in light of the repeal of the child tax credit expected to lead to an increase in the percentage of children living in poverty. The early findings of enhanced neurodevelopment in the innovative “Baby's first years study” that provides monthly supplemental income (as little as $330/month) to high-risk mother-infant dyads (Troller-Renfree et al., 2022) provides further evidence of the critical need to provide financial support in families with young children. What remains to be determined from this landmark study is how income focused interventions influence the range of health metrics and biomarkers of health risk. In the interim, the established relation between TL and most of the leading causes of morbidity and mortality, when combined with our finding of an association, in childhood, between income-based SES and TL, should sound an alarm bell for those invested in healthy child development. Continued inequity in income, across all racial groups, can only be expected to drive poor health trajectories for socioeconomically disadvantaged families and greater health disparities. Achieving health equity requires addressing key modifiable factors driving them and clearly needs to begin early in life and be centered within the family and community for greatest effectiveness.

Funding

Funding was supported by the Telomere Research Network, NIH Project 5U24AG066528 Drury, PI. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript. We would like to thank Amy Coder and the Meta-Analysis Systematic Review Support (MARS) Program at Tulane University for their assistance throughout the development and preparation of this publication. Any questions regarding the review protocol can be directed to Stacy S. Drury, MD, PhD. Data collection and extraction forms, data used for analyses, and analytic code are accessible upon request from Stacy S. Drury, MD, PhD. The authors report no financial conflict of interest.

Footnotes

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.ssmph.2023.101380.

Appendix A. Supplementary data

The following are the Supplementary data to this article.

Multimedia component 1
mmc1.pdf (131.7KB, pdf)

figs1.

figs1

Open in a new tab

figs2.

figs2

Open in a new tab

Data availability

Data will be made available on request.

References

  1. Asok A., et al. Parental responsiveness moderates the association between early-life stress and reduced telomere length. Development and Psychopathology. 2013;25(3):577–585. doi: 10.1017/S0954579413000011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Assari S. Family socioeconomic status and exposure to childhood trauma: Racial differences. Children. 2020;7(6) doi: 10.3390/children7060057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Assari S., et al. Minorities' diminished returns of parental educational attainment on adolescents' social, emotional, and behavioral problems. Children. 2020;7(5):49. doi: 10.3390/children7050049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bailey Z.D., et al. Structural racism and health inequities in the USA: Evidence and interventions. Lancet. 2017;389(10077):1453–1463. doi: 10.1016/S0140-6736(17)30569-X. [DOI] [PubMed] [Google Scholar]
  5. Barraclough J., et al. Early and late childhood telomere length predict subclinical atherosclerosis at age 14 yrs. - the CardioCAPS study. International Journal of Cardiology. 2019;278:250–253. doi: 10.1016/j.ijcard.2018.12.065. [DOI] [PubMed] [Google Scholar]
  6. Baskind M.J., et al. Obesity at age 6 Months is associated with shorter preschool leukocyte telomere length independent of parental telomere length. The Journal of Pediatrics. 2021;233:141–149. doi: 10.1016/j.jpeds.2020.11.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Baskind M.J., et al. Obesity at age 6 Months is associated with shorter preschool leukocyte telomere length independent of parental telomere length. The Journal of Pediatrics. 2021 doi: 10.1016/j.jpeds.2020.11.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Beebe-Dimmer J. Childhood and adult socioeconomic conditions and 31-year mortality risk in women. American Journal of Epidemiology. 2004;159(5):481–490. doi: 10.1093/aje/kwh057. [DOI] [PubMed] [Google Scholar]
  9. Berenson G.S., et al. Atherosclerosis of the aorta and coronary arteries and cardiovascular risk factors in persons aged 6 to 30 years and studied at necropsy (the Bogalusa Heart Study) The American Journal of Cardiology. 1992;70(9):851–858. doi: 10.1016/0002-9149(92)90726-f. [DOI] [PubMed] [Google Scholar]
  10. Bianco-Miotto T., et al. Epigenetics and DOHaD: From basics to birth and beyond. Journal of Developmental Origins of Health and Disease. 2017;8(5):513–519. doi: 10.1017/S2040174417000733. [DOI] [PubMed] [Google Scholar]
  11. Black/african American - the office of minority health. 2021. [Google Scholar]
  12. Bosquet Enlow M., et al. Patterns of change in telomere length over the first three years of life in healthy children. Psychoneuroendocrinology. 2020;115 doi: 10.1016/j.psyneuen.2020.104602. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Bountziouka V., et al. Modifiable traits, healthy behaviours, and leukocyte telomere length: A population-based study in UK biobank. The Lancet Healthy Longevity. 2022;3(5):e321–e331. doi: 10.1016/S2666-7568(22)00072-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Braveman P. Healt disparities and health equity: Concepts and measurement. Annual Review of Public Health. 2006;27(1):167–194. doi: 10.1146/annurev.publhealth.27.021405.102103. [DOI] [PubMed] [Google Scholar]
  15. Braveman P.A., et al. Socioeconomic status in health research. JAMA. 2005;294(22):2879. doi: 10.1001/jama.294.22.2879. [DOI] [PubMed] [Google Scholar]
  16. Broer L., et al. Meta-analysis of telomere length in 19 713 subjects reveals high heritability, stronger maternal inheritance and a paternal age effect. European Journal of Human Genetics. 2013;21(10):1163–1168. doi: 10.1038/ejhg.2012.303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Brown L., Needham B., Ailshire J. Telomere length among older U.S. Adults: Differences by race/ethnicity, gender, and age. Journal of Aging and Health. 2017;29(8):1350–1366. doi: 10.1177/0898264316661390. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Carroll J.E., et al. Prenatal maternal stress prospectively relates to shorter child buccal cell telomere length. Psychoneuroendocrinology. 2020:121. doi: 10.1016/j.psyneuen.2020.104841. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Chae D., et al. Discrimination, racial bias and telomere length in African-American men. American Journal of Preventive Medicine. 2014;46(2):103–111. doi: 10.1016/j.amepre.2013.10.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Chae D., et al. Discrimination, mental health, and leukocyte telomere length among African American men. Psychoneuroendocrinology. 2016;63:10–16. doi: 10.1016/j.psyneuen.2015.09.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Chen L., et al. Higher chocolate intake is associated with longer telomere length among adolescents. Pediatric Research. 2020;87(3):602–607. doi: 10.1038/s41390-019-0590-6. [DOI] [PubMed] [Google Scholar]
  22. Chetty R., et al. The association between income and life expectancy in the United States, 2001-2014. JAMA. 2016;315(16):1750. doi: 10.1001/jama.2016.4226. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Cohen S., et al. Childhood socioeconomic status and adult health. Annals of the New York Academy of Sciences. 2010;1186(1):37–55. doi: 10.1111/j.1749-6632.2009.05334.x. [DOI] [PubMed] [Google Scholar]
  24. Covidence systematic review software, veritas health innovation, melbourne, Australia. www.covidence.org Available at:
  25. David B. Goesling, and gerald, socioeconomic status, school quality, and national economic development: A cross‐national Analysis of the “heyneman‐loxley effect” on Mathematics and science achievement. Comparative Education Review. 2002;46(3):291–312. [Google Scholar]
  26. Diez Roux A.V., et al. Race/ethnicity and telomere length in the multi-ethnic study of atherosclerosis. Aging Cell. 2009;8(3):251–257. doi: 10.1111/j.1474-9726.2009.00470.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Dismukes A.R., et al. Diurnal and stress-reactive dehydroepiandrosterone levels and telomere length in youth. Endocrine Connections. 2016;5(3):107–114. doi: 10.1530/EC-16-0007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Drury S.S., et al. The association of telomere length with family violence and disruption. Pediatrics. 2014;134(1):E128–E137. doi: 10.1542/peds.2013-3415. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Drury S.S., et al. Growing up or growing old? Cellular aging linked with testosterone reactivity to stress in youth. The American Journal of the Medical Sciences. 2014;348(2):92–100. doi: 10.1097/MAJ.0000000000000299. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Drury S., et al. Setting the trajectory: Racial disparities in newborn telomere length. The Journal of Pediatrics. 2015;166(5):1181–1186. doi: 10.1016/j.jpeds.2015.01.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Drury S.S., et al. Setting the trajectory: Racial disparities in newborn telomere length. The Journal of Pediatrics. 2015;166(5):1181–1186. doi: 10.1016/j.jpeds.2015.01.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Duncan G.J., et al. Optimal indicators of socioeconomic status for health research. American Journal of Public Health. 2002;92(7):1151–1157. doi: 10.2105/ajph.92.7.1151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Entringer S., et al. Maternal estriol concentrations in early gestation predict infant telomere length. The Journal of Cinical Endocrinology and Metabolism. 2015;100(1):267–273. doi: 10.1210/jc.2014-2744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Epel E.S., et al. Accelerated telomere shortening in response to life stress. Proceedings of the National Academy of Sciences. 2004;101(49):17312–17315. doi: 10.1073/pnas.0407162101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Esc package - RDocumentation. 2021. https://www.rdocumentation.org/packages/esc/versions/0.5.1 Available from: [Google Scholar]
  36. Esteves K.C., et al. Adverse childhood experiences: Implications for offspring telomere length and psychopathology. American Journal of Psychiatry. 2020;177(1):47–57. doi: 10.1176/appi.ajp.2019.18030335. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Etzel L., et al. Psychoneuroendocrinology; 2020. Intergenerational transmission of childhood trauma? Testing cellular aging in mothers exposed to sexual abuse and their children. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Galobardes B. Childhood socioeconomic circumstances and cause-specific mortality in adulthood: Systematic review and interpretation. Epidemiologic Reviews. 2004;26(1):7–21. doi: 10.1093/epirev/mxh008. [DOI] [PubMed] [Google Scholar]
  39. Galobardes B., Lynch J., Smith G.D. Measuring socioeconomic position in health research. British Medical Bulletin. 2007;81–82(1):21–37. doi: 10.1093/bmb/ldm001. [DOI] [PubMed] [Google Scholar]
  40. Gardner M., et al. Gender and telomere length: Systematic review and meta-analysis. Experimental Gerontology. 2014;51:15–27. doi: 10.1016/j.exger.2013.12.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Geronimus A. The weathering hypothesis and the health of african-American women and infants: Evidence and speculations. Ethnicity & Disease. 1992;2(3):207–221. [PubMed] [Google Scholar]
  42. Geyer S. Education, income, and occupational class cannot be used interchangeably in social epidemiology. Empirical evidence against a common practice. Journal of Epidemiology & Community Health. 2006;60(9):804–810. doi: 10.1136/jech.2005.041319. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Global inequality: How the U.S. compares. 2021. [Google Scholar]
  44. Gluckman P.D., Hanson M.A., Buklijas T. A conceptual framework for the developmental origins of health and disease. Journal of Developmental Origins of Health and Disease. 2010;1(1):6–18. doi: 10.1017/S2040174409990171. [DOI] [PubMed] [Google Scholar]
  45. Guarneri‐White M.E., et al. It's more than skin‐deep: The relationship between social victimization and telomere length in adolescence. Aggressive Behavior. 2018;44(4):337–347. doi: 10.1002/ab.21755. [DOI] [PubMed] [Google Scholar]
  46. Hanssen L.M., et al. The relationship between childhood psychosocial stressor level and telomere length: A meta-analysis. Health Psychology Research. 2017;5(1) doi: 10.4081/hpr.2017.6378. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Haycock P.C., et al. Leucocyte telomere length and risk of cardiovascular disease: Systematic review and meta-analysis. BMJ. 2014;349:g4227. doi: 10.1136/bmj.g4227. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. He J., et al. Risk factors for congestive heart failure in US men and women. Archives of Internal Medicine. 2001;161(7):996. doi: 10.1001/archinte.161.7.996. [DOI] [PubMed] [Google Scholar]
  49. Henje Blom E., et al. Peripheral telomere length and hippocampal volume in adolescents with major depressive disorder. Translational Psychiatry. 2015;5(11) doi: 10.1038/tp.2015.172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Huedo-Medina T.B., et al. Assessing heterogeneity in meta-analysis: Q statistic or I2 index? Psychological Methods. 2006;11(2):193–206. doi: 10.1037/1082-989X.11.2.193. [DOI] [PubMed] [Google Scholar]
  51. Humphreys K.L., et al. Depressive symptoms predict change in telomere length and mitochondrial DNA copy number across adolescence. Journal of the American Academy of Child & Adolescent Psychiatry. 2020;59(12):1364–1370. doi: 10.1016/j.jaac.2019.09.031. e2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. James S., et al. Sleep duration and telomere length in children. The Journal of Pediatrics. 2017;187:247–252. doi: 10.1016/j.jpeds.2017.05.014. e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Jiang Y.P., et al. Basal cortisol, cortisol reactivity, and telomere length: A systematic review and meta-analysis. Psychoneuroendocrinology. 2019;103:163–172. doi: 10.1016/j.psyneuen.2019.01.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Jones C., et al. Differences in placental telomere length suggest a link between racial disparities in birth outcomes and cellular aging. American Journal of Obstetrics and Gynecology. 2017;216(3):294e1–294e8. doi: 10.1016/j.ajog.2016.11.1027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Juster R.P., McEwen B.S., Lupien S.J. Allostatic load biomarkers of chronic stress and impact on health and cognition. Neuroscience & Biobehavioral Reviews. 2010;35(1):2–16. doi: 10.1016/j.neubiorev.2009.10.002. [DOI] [PubMed] [Google Scholar]
  56. Kachmar A.G., et al. Socioeconomic status in pediatric health research: A scoping review. The Journal of Pediatrics. 2019;213:163–170. doi: 10.1016/j.jpeds.2019.06.005. [DOI] [PubMed] [Google Scholar]
  57. Kastner M., et al. Age-specific search strategies for medline. Journal of Medical Internet Research. 2006;8(4):e25. doi: 10.2196/jmir.8.4.e25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Khan R., et al. Association of childhood socioeconomic status with leukocyte telomere length among african Americans and the mediating role of behavioral and psychosocial factors: Results from the GENE-FORECAST study. Journal of Racial and Ethnic Health Disparities. 2022;9(3):1012–1023. doi: 10.1007/s40615-021-01040-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Kjaer T.W., et al. Shorter preschool, leukocyte telomere length is associated with obesity at age 9 in Latino children. Clinical Obesity. 2018;8(2):88–94. doi: 10.1111/cob.12233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Kjaer T.W., Medrano R., Wojcicki J.M. Rapid infant weight gain and prenatal maternal depression are associated with prehypertension/hypertension before age 9. Journal of Human Hypertension. 2020;34(11):795–801. doi: 10.1038/s41371-020-0306-8. [DOI] [PubMed] [Google Scholar]
  61. Kliewer W., Robins J.L. Adverse childhood experiences are associated with cardiometabolic risk indicators and telomere length in low-income african-american adolescents. International Journal of Behavioral Medicine. 2021 doi: 10.1007/s12529-021-09978-w. [DOI] [PubMed] [Google Scholar]
  62. Koss K.J., et al. Early puberty and telomere length in preadolescent girls and mothers. The Journal of Pediatrics. 2020;222:193–199. doi: 10.1016/j.jpeds.2020.02.075. e5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Kroenke C.H., et al. Autonomic and adrenocortical reactivity and buccal cell telomere length in kindergarten children. Psychosomatic Medicine. 2011;73(7):533–540. doi: 10.1097/PSY.0b013e318229acfc. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Kuh D. Mortality in adults aged 26-54 years related to socioeconomic conditions in childhood and adulthood: Post war birth cohort study. BMJ. 2002;325(7372):1076–1080. doi: 10.1136/bmj.325.7372.1076. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Kuh D. Life course epidemiology. Journal of Epidemiology & Community Health. 2003;57(10):778–783. doi: 10.1136/jech.57.10.778. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Lantz P.M., et al. Socioeconomic factors, health behaviors, and mortality. JAMA. 1998;279(21):1703. doi: 10.1001/jama.279.21.1703. [DOI] [PubMed] [Google Scholar]
  67. Lee E.Y., et al. Ambient air pollution, asthma drug response, and telomere length in African American youth. Journal of Allergy and Clinical Immunology. 2019;144(3):839–845. doi: 10.1016/j.jaci.2019.06.009. e10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Lin L., Chu H. Quantifying publication bias in meta-analysis. Biometrics. 2018;74(3):785–794. doi: 10.1111/biom.12817. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Lindrose A.R., Drury S. 2020. Minimum reporting recommendations for PCR-based telomere length measurement. [Google Scholar]
  70. Lin J., et al. Telomere length measurement by qPCR–Summary of critical factors and recommendations for assay design. Psychoneuroendocrinology. 2019;99:271–278. doi: 10.1016/j.psyneuen.2018.10.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Luby J.L., et al. Neurodevelopmental optimization after early-life adversity: Cross-species studies to elucidate sensitive periods and brain mechanisms to inform early intervention. Trends in Neurosciences. 2020;43(10):744–751. doi: 10.1016/j.tins.2020.08.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Manczak E.M., Gotlib I.H. Relational victimization and telomere length in adolescent girls. Journal of Research on Adolescence. 2020;30(Suppl 1):39–45. doi: 10.1111/jora.12447. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. McEwen C.A., McEwen B.S. Social structure, adversity, toxic stress, and intergenerational poverty: An early childhood model. Annual Review of Sociology. 2017;43(1):445–472. [Google Scholar]
  74. Merkin S.S., et al. Neighborhoods and cumulative biological risk profiles by race/ethnicity in a national sample of U.S. Adults: NHANES III. Annals of Epidemiology. 2009;19(3):194–201. doi: 10.1016/j.annepidem.2008.12.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Merrill L.C., et al. The differential impact of oxytocin receptor gene in violence-exposed boys and girls. International Journal of Developmental Neuroscience. 2017;59:60–67. doi: 10.1016/j.ijdevneu.2017.03.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Miller J.G., et al. Early life stress, frontoamygdala connectivity, and biological aging in adolescence: A longitudinal investigation. Cerebral Cortex. 2020;30(7):4269–4280. doi: 10.1093/cercor/bhaa057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Miri M., et al. Air pollution and telomere length in adults: A systematic review and meta-analysis of observational studies. Environmental Pollution. 2019;244:636–647. doi: 10.1016/j.envpol.2018.09.130. [DOI] [PubMed] [Google Scholar]
  78. Mitchell C., et al. Social disadvantage, genetic sensitivity, and children's telomere length. Proceedings of the National Academy of Sciences. 2014;111(16):5944–5949. doi: 10.1073/pnas.1404293111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  79. Mitchell C., et al. Father loss and child telomere length. Pediatrics. 2017;140(2) doi: 10.1542/peds.2016-3245. [DOI] [PMC free article] [PubMed] [Google Scholar]
  80. Moola S., et al. JBI; 2020. Chapter 7: Systematic reviews of etiology and risk JBI manual for evidence synthesis. [Google Scholar]
  81. Moyzis R.K., et al. A highly conserved repetitive DNA sequence, (TTAGGG)n, present at the telomeres of human chromosomes. Proceedings of the National Academy of Sciences. 1988;85(18):6622–6626. doi: 10.1073/pnas.85.18.6622. [DOI] [PMC free article] [PubMed] [Google Scholar]
  82. National income - net national income - OECD Data. 2021. http://data.oecd.org/natincome/net-national-income.htm Available from: [Google Scholar]
  83. Needham B.L., et al. Socioeconomic status and cell aging in children. Social Science & Medicine. 2012;74(12):1948–1951. doi: 10.1016/j.socscimed.2012.02.019. [DOI] [PubMed] [Google Scholar]
  84. Needham B.L., et al. Socioeconomic status, health behavior, and leukocyte telomere length in the National Health and Nutrition Examination Survey, 1999-2002. Social Science & Medicine. 2013;85:1–8. doi: 10.1016/j.socscimed.2013.02.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  85. Nelson B.W., Allen N.B., Laurent H. Infant HPA axis as a potential mechanism linking maternal mental health and infant telomere length. Psychoneuroendocrinology. 2018;88:38–46. doi: 10.1016/j.psyneuen.2017.11.008. [DOI] [PubMed] [Google Scholar]
  86. Nelson B.W., et al. Maternal stress and social support prospectively predict infant inflammation. Brain, Behavior, and Immunity. 2020;86:14–21. doi: 10.1016/j.bbi.2019.05.010. [DOI] [PubMed] [Google Scholar]
  87. Nelson B.W., et al. Psychobiological markers of allostatic load in depressed and nondepressed mothers and their adolescent offspring. Journal of Child Psychology and Psychiatry. 2021;62(2):199–211. doi: 10.1111/jcpp.13264. [DOI] [PMC free article] [PubMed] [Google Scholar]
  88. Oliveira B.S., et al. Systematic review of the association between chronic social stress and telomere length: A life course perspective. Ageing Research Reviews. 2016;26:37–52. doi: 10.1016/j.arr.2015.12.006. [DOI] [PubMed] [Google Scholar]
  89. Pantesco E.J., et al. Multiple forms of discrimination, social status, and telomere length: Interactions within race. Psychoneuroendocrinology. 2018;98:119–126. doi: 10.1016/j.psyneuen.2018.08.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  90. Puterman E., et al. Lifespan adversity and later adulthood telomere length in the nationally representative US Health and Retirement Study. Proceedings of the National Academy of Sciences of the United States of America. 2016;113(42):E6335–E6342. doi: 10.1073/pnas.1525602113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  91. R: The R project for statistical computing. 2021. https://www.r-project.org/ Available from: [Google Scholar]
  92. Ridout K.K., et al. Early life adversity and telomere length: A meta-analysis. Molecular Psychiatry. 2018;23(4):858–871. doi: 10.1038/mp.2017.26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  93. Ridout K.K., et al. Childhood maltreatment, behavioral adjustment, and molecular markers of cellular aging in preschool-aged children: A cohort study. Psychoneuroendocrinology. 2019;107:261–269. doi: 10.1016/j.psyneuen.2019.05.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  94. Robbins J., et al. Socioeconomic status and type 2 diabetes in african American and non-hispanic white women and men: Evidence from the third national health and nutrition examination survey. American Journal of Public Health. 2001;91(1):76–83. doi: 10.2105/ajph.91.1.76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  95. Robertson T., et al. Is socioeconomic status associated with biological aging as measured by telomere length? Epidemiologic Reviews. 2013;35(1):98–111. doi: 10.1093/epirev/mxs001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  96. Robles T.F., et al. Emotions and family interactions in childhood: Associations with leukocyte telomere length emotions, family interactions, and telomere length. Psychoneuroendocrinology. 2016;63:343–350. doi: 10.1016/j.psyneuen.2015.10.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  97. Schmeer K.K., Tarrence J. Racial-ethnic disparities in inflammation: Evidence of weathering in childhood? Journal of Health and Social Behavior. 2018;59(3):411–428. doi: 10.1177/0022146518784592. [DOI] [PMC free article] [PubMed] [Google Scholar]
  98. Schutte N., Malouff J. Stress and Health; 2014. The relationship between perceived stress and telomere length: A meta-analysis. [DOI] [PubMed] [Google Scholar]
  99. Shavers V.L. Measurement of socioeconomic status in health disparities research. Journal of the National Medical Association. 2007;99(9):1013–1023. [PMC free article] [PubMed] [Google Scholar]
  100. Skinner M.L., et al. Allostasis model facilitates understanding race differences in the diurnal cortisol rhythm. Development and Psychopathology. 2011;23(4):1167–1186. doi: 10.1017/S095457941100054X. [DOI] [PMC free article] [PubMed] [Google Scholar]
  101. Sosnowski D.W., et al. The association between adverse childhood experiences and child telomere length: Examining self-regulation as a behavioral mediator. Child Development. 2021;92(2):746–759. doi: 10.1111/cdev.13441. [DOI] [PubMed] [Google Scholar]
  102. Theall K., Drury S., Shirtcliff E. Cumulative neighborhood risk of psychosocial stress and allostatic load in adolescents. American Journal of Epidemiology. 2012;176(Suppl 7):S164–S174. doi: 10.1093/aje/kws185. [DOI] [PMC free article] [PubMed] [Google Scholar]
  103. Theall K.P., Drury S., Shirtcliff E.A. Shortened average telomere length in children and neighborhood disorder: Connecting community level stress and cellular response. American Journal of Epidemiology. 2012;175:S93. [Google Scholar]
  104. Theall K.P., et al. Neighborhood disorder and telomeres: Connecting children's exposure to community level stress and cellular response. Social Science & Medicine. 2013;85:50–58. doi: 10.1016/j.socscimed.2013.02.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  105. Theall K.P., et al. Association between neighborhood violence and biological stress in children. JAMA Pediatrics. 2017;171(1):53–60. doi: 10.1001/jamapediatrics.2016.2321. [DOI] [PMC free article] [PubMed] [Google Scholar]
  106. Theall K.P., et al. Childhood obesity and the associated roles of neighborhood and biologic stress. Prev Med Rep. 2019;14 doi: 10.1016/j.pmedr.2019.100849. [DOI] [PMC free article] [PubMed] [Google Scholar]
  107. Troller-Renfree S.V., et al. The impact of a poverty reduction intervention on infant brain activity. Proceedings of the National Academy of Sciences. 2022;119(5) doi: 10.1073/pnas.2115649119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  108. Viechtbauer W. 2010. Conducting meta-analyses in R with the metafor package.https://www.jstatsoft.org/index.php/jss [Google Scholar]
  109. Willeit P., et al. Leucocyte telomere length and risk of type 2 diabetes mellitus: New prospective cohort study and literature-based meta-analysis. PLoS One. 2014;9(11) doi: 10.1371/journal.pone.0112483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  110. Williams D.R., et al. Racial differences in physical and mental health. Journal of Health Psychology. 1997;2(3):335–351. doi: 10.1177/135910539700200305. [DOI] [PubMed] [Google Scholar]
  111. Williams D.R., Priest N., Anderson N. Understanding associations between race, socioeconomic status and health: Patterns and prospects. Health Psychology. 2016;35(4):407–411. doi: 10.1037/hea0000242. [DOI] [PMC free article] [PubMed] [Google Scholar]
  112. Wojcicki J.M., et al. Early exclusive breastfeeding is associated with longer telomeres in Latino preschool children. American Journal of Clinical Nutrition. 2016;104(2):397–405. doi: 10.3945/ajcn.115.115428. [DOI] [PMC free article] [PubMed] [Google Scholar]
  113. Wojcicki J.M., et al. Increased cellular aging by 3 Years of age in latino, preschool children who consume more sugar-sweetened beverages: A pilot study. Childhood Obesity. 2018;14(3):149–157. doi: 10.1089/chi.2017.0159. [DOI] [PMC free article] [PubMed] [Google Scholar]
  114. Worthy l., Lavigne T., Romero F. 2020. Socioeconomic status (SES) [Google Scholar]
  115. Zeiger A., et al. Genetic determinants of telomere length in african American youth. Scientific Reports. 2018;8(1) doi: 10.1038/s41598-018-31238-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  116. Zhu H., et al. Leukocyte telomere length in healthy caucasian and african-American adolescents: Relationships with race, sex, adiposity, adipokines, and physical activity. The Journal of Pediatrics. 2011;158(2):215–220. doi: 10.1016/j.jpeds.2010.08.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  117. Zhu H., et al. High sodium intake is associated with short leukocyte telomere length in overweight and obese adolescents. Circulation. 2014:129. doi: 10.1038/ijo.2015.51. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Multimedia component 1
mmc1.pdf (131.7KB, pdf)

Data Availability Statement

Data will be made available on request.

Articles from SSM - Population Health are provided here courtesy of Elsevier

RESOURCES