Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jan 1.
Published in final edited form as: Soc Sci Med. 2020 Dec 8;269:113592. doi: 10.1016/j.socscimed.2020.113592

Early origins of socioeconomic inequalities in chronic inflammation: Evaluating the contributions of low birth weight and short breastfeeding

Thomas W McDade 1,2,3, Stephanie M Koning 2
PMCID: PMC7780588  NIHMSID: NIHMS1653471  PMID: 33360022

1. Introduction

Wide socioeconomic gaps in morbidity and mortality in the US are well-established, but their origins are not. Lifestyle factors and health behaviors in adulthood, as well as access to health care services, explain only a portion of social gradients in health (Woolf et al. 2007, Berkman et al. 2014). Psychosocial stressors in adulthood are also important (Adler and Snibbe 2003, Jackson et al. 2010, Goosby et al. 2018), but it has become increasingly apparent that contexts and experiences earlier in development have direct, as well as indirect, effects on physiological systems and later life morbidity (Hayward and Gorman 2004, Gluckman et al. 2008, Goosby et al. 2016). We are therefore challenged to consider the extent to which socioeconomic inequalities in adult health trace back—at least in part—to socially patterned exposures earlier in the life course.

In this analysis we focus on chronic inflammation as a plausible physiological pathway through which environments early in life shape trajectories of health in adulthood. Higher levels of chronic inflammation—indicated here by C-reactive protein (CRP)—predict increased risk for all-cause mortality, as well as risk for cardiometabolic diseases that are major public health burdens in the US (Harris et al. 1999, Pradhan et al. 2001, Pearson et al. 2003, Kuo et al. 2005, Kuo et al. 2006). We use data from a large, nationally representative sample of young adults with contextual, behavioral, and physiological measures collected at multiple time points. While rates of clinical disease are low in young adulthood, trajectories of later life health are well established and can be revealed by biological measures such as CRP.

After briefly outlining the rationale for considering the developmental origins of social inequalities in adult health, we discuss the life course determinants of chronic inflammation and highlight birth weight and breastfeeding duration as particularly important aspects of the prenatal and postnatal environment. We describe the socioeconomic distribution of breastfeeding duration and birth weight, and note that shorter durations of breastfeeding and lower birth weight deliveries are strongly associated with lower socioeconomic status (SES) in the US. We therefore hypothesize that SES gradients in chronic inflammation are a partial product of breastfeeding duration and birth weight, and we outline a structural equation modeling framework for testing this hypothesis.

1.2. Developmental origins of social inequalities in health

Developmental, life course approaches highlight the importance of contexts and experiences early in life as potential determinants of later life health, independent of circumstances in adulthood (Barker et al. 2002, Hayward and Gorman 2004, Kittleson et al. 2006, Gluckman et al. 2008, Shonkoff et al. 2009, Halfon et al. 2014). For example, the onset of coronary heart disease (CHD) has been investigated among graduates of the Johns Hopkins medical school, and even for this group of highly educated, affluent physicians, low SES in childhood more than doubles the risk of CHD by age 50 (Kittleson et al. 2006). More broadly, a growing body of research in biodemography and life course epidemiology has established the “long arm of childhood,” with early life circumstances shaping health outcomes decades later through several direct and indirect pathways (Preston et al. 1998, Blackwell et al. 2001, Hayward and Gorman 2004, Case and Paxson 2010).

The biological processes that link environments in infancy with health in adulthood are the primary focus of research on the developmental origins of health and disease (DOHaD). Building on early observations linking socioeconomic conditions around the time of birth and life expectancy within birth cohorts (Kermack et al. 1934, Forsdahl 1977), scholars across a wide range of disciplines have applied observational, quasi-experimental, and experimental animal models to illuminate the mechanisms through which early environments can have disproportionate and lasting effects on the function of key physiological systems (Barker et al. 2002, Gluckman et al. 2008, Gluckman et al. 2016). Initial analyses, led primarily by Barker and colleagues, focused on birthweight as an indicator of nutrient transfer to the fetus during gestation, with lower birth weight predicting increased cardiovascular disease risk (Barker et al. 1989, Barker 1998, Barker et al. 2006). Subsequent research has considered a wider range of prenatal and postnatal factors that predict increased risk for morbidity and mortality through modifications to physiological systems, organ growth, and metabolism during sensitive periods of development (Cameron and Demerath 2002, Gluckman et al. 2008, Bengtsson and Broström 2009, Smith and Ryckman 2015).

While the DOHaD framework emphasizes the biological mechanisms through which developmental contexts shape risk for disease later in life, it generally does not attend to the unequal distribution of these contexts and how they may, in turn, contribute to social inequalities in morbidity and mortality in adulthood. The concepts of “embodiment” and “biological embedding” have provided opportunities for synthesis: Both underscore environmental sensitivity and developmental plasticity as defining features of human biology, and both emphasize that socially patterned contexts and experiences have enduring effects on the phenotype that shape risk for a wide range of health outcomes (Gravlee 2009, Kuzawa and Sweet 2009, Hertzman and Boyce 2010, Thayer and Kuzawa 2013, Hoke and McDade 2014). In the analyses below we focus on birth weight and breastfeeding as important aspects of the developmental environment that influence chronic inflammation and trajectories of cardiometabolic disease risk. However, we go beyond a typical DOHaD approach in considering how these environments are patterned by SES and their potential role as drivers of social inequalities in chronic inflammation in adulthood.

1.3. Life course determinants of chronic inflammation

When the immune system detects injury or an invading pathogen, an inflammatory response mobilizes cellular and biochemical defenses that repair damaged tissue and eliminate the pathogen. This process is coordinated by pro- (e.g., IL1b, IL6, TNFa) and anti-inflammatory (e.g., IL10) cytokines, which—along with CRP, an acute phase reactant produced by the liver—are frequently measured in circulation as indicators of ongoing inflammatory activity (Pearson et al. 2003, Black et al. 2004). Effective regulation of inflammation is essential since excessive or chronic activation can cause tissue damage and contribute to the development of a wide range of negative health outcomes, including CHD, type 2 diabetes, metabolic syndrome, late-life disability, and mortality (Pradhan et al. 2001, Ridker et al. 2003, Kuo et al. 2006, Jenny et al. 2007). Similarly, there is growing awareness that inflammatory processes during pregnancy play important roles in adverse birth outcomes, including preterm delivery and intra-uterine growth restriction (Romero et al. 2007, Challis et al. 2009).

Socioeconomic inequalities in these health outcomes are widely recognized (Adler et al. 1994, Mackenbach and Kunst 1997, Marmot et al. 1997, Keppel et al. 2002, Cohen et al. 2003), and inflammation has become a major focus of research as a potentially important mechanism linking social environments and health (Alley et al. 2006, Crimmins and Finch 2006, Nazmi and Victora 2007, Ranjit et al. 2007). Several studies have demonstrated socioeconomic gradients in CRP, with higher concentrations of CRP found in groups with lower levels of education and income (Alley et al. 2006, Nazmi and Victora 2007, Ranjit et al. 2007). Complementary lines of inquiry consider whether inflammation mediates the impact of intermediate psychosocial factors (e.g., stress, depression, social isolation) on morbidity and mortality (Miller et al. 2002, Miller et al. 2003, McDade et al. 2006, Cole et al. 2007, Ranjit et al. 2007). Prior research has suggested that these factors, as well as obesity/overweight and infectious exposures, represent more proximate factors that explain part—but not all—of the established link between SES and inflammation (Ford et al. 2004, Aiello and Kaplan 2009, McDade et al. 2010).

Although inflammation is most often studied in relation to environments in adulthood, there is growing evidence that early life environments have long-term, independent effects on immune regulation and inflammation (Shanks and Lightman 2001, McDade 2006, Taylor 2010). For example, low early life SES is associated with elevated CRP among adults in the CARDIA study (Taylor et al. 2006), and it predicts increased pro-inflammatory and decreased anti-inflammatory gene expression in otherwise healthy adults (Miller et al. 2009). Low birth weight has also been associated with elevated inflammation in adulthood in several cohorts (Sattar et al. 2004, Danese et al. 2007, Tzoulaki et al. 2008, McDade et al. 2010).

In a recent analysis of Add Health data, Goosby and colleagues (Goosby et al. 2016) document “chains of risk” in which birthweight and body mass early in life predict body mass in adolescence and young adulthood, which in turn predict chronic inflammation. McDade and colleagues have also reported significant associations between birth weight and CRP in young adulthood in Add Health (McDade et al. 2014). Of note, a one kilogram increase in birth weight predicted a 9.2% reduction in adult CRP across the cohort, whereas sibling comparison models reported stronger associations, with a 24% reduction in CRP for each kilogram. Results from sibling comparisons are significant in that they control for many hard-to-measure attributes of families and communities that might otherwise confound associations between early environments and later life health, and they increase confidence in concluding that aspects of the prenatal environment have causal effects on the regulation of inflammation in adulthood.

While birth weight has been the primary focus in DOHaD, the postnatal nutritional environment can also have lasting effects on inflammation. Breastfeeding, in particular, provides nutritional and immunological support to infants, and studies in New Zealand, Scotland, and the US have documented lower CRP in adults who were breastfed as infants (Williams et al. 2006, Rudnicka et al. 2007, McDade et al. 2014, Olson and Hayward 2017). The effects of breastfeeding on inflammation can be direct, through the shaping of regulatory pathways during sensitive periods of immune development in infancy (Field 2005, McDade 2012), and they can be indirect, through programming of metabolic pathways that contribute to the accumulation of body fat and the emergence of overweight/obesity (Harder et al. 2005, Schaffler et al. 2006). Emerging evidence also points to obesogenic effects of formula feeding on the gut microbiota in infancy (Forbes et al. 2018).

In Add Health, we previously documented a negative association between breastfeeding duration and adult CRP, with an important inflection point at 3 months: CRP concentrations in young adulthood were almost 30% higher among individuals who were not breastfed as infants in comparison with those breastfed for 3 months or longer (McDade et al. 2014). Sibling comparisons suggested the possibility of stronger effects, with 46% higher CRP in siblings breastfed for less than three months (or not at all), in comparison with a sister for brother who was breastfed for longer than three months. Mediation analyses suggested that approximately 40% of the association between breastfeeding duration and adult CRP in Add Health was accounted for by waist circumference, consistent with prior studies reporting associations between breastfeeding and adiposity in Add Health and other cohorts (Harder et al. 2005, Metzger and McDade 2010, Horta et al. 2015, Olson and Hayward 2017), and between adiposity and chronic inflammation (Schaffler et al. 2006).

1.4. Social inequalities in birth outcomes and breastfeeding duration

Race- and class-based inequalities in birth outcomes in the US are large and growing: Infants born to Black women are almost twice as likely as White infants to be born preterm or low birth weight, and more than twice as likely to die in infancy (Mathews and MacDorman 2013, Martin et al. 2015). Similar trends are seen when comparing low income or poorly educated women with high SES women: Risk of infant mortality is more than double for a high school drop out compared with a college graduate (Blumenshine et al. 2010, Green and Hamilton 2019). And after decades of overall improvement, average birth weights in the US may now be on the decline, with steeper drops for low SES and Black women that are widening birth weight inequalities (Morisaki et al. 2013, Catov et al. 2015). This is a significant, and costly, public health problem: Preterm deliveries alone cost more than $26 billion annually (Statistics 2012).

Patterns of breastfeeding initiation and duration follow similar social gradients in the US, with overall low levels of breastfeeding in comparison with current guidelines. The American Academy of Pediatrics (AAP) recommends exclusive breastfeeding—defined as feeding an infant only breast milk, with no other liquids or solids other than vitamins or medicines—for six months, with continued breastfeeding for one year or longer as complementary foods are introduced (Breastfeeding 2012). As of 2016, 83.8% of infants born in the US were breastfed in some form, but only 25.4% were exclusively breastfed for six months, and 36.2% were breastfed at one year (Prevention 2020). While these rates fall short of current guidelines, they are much better than the late 70s and early 80s, when breastfeeding was initiated for only one in two infants in the US, and when fewer than one in four were still receiving any breastmilk at six months (Medicine et al. 1991, Ryan 1997).

As with birth weight, social inequalities in breastfeeding are evident. Babies born to college graduates are approximately twice as likely to meet AAP recommendations as high school graduates (32.9 vs. 18.7% exclusive breastfeeding to 6 months; 48.6 vs. 24.0% breastfeeding at 12 months). Similarly, poverty-income-ratio (PIR) predicts breastfeeding, with 31.0% of the most affluent households (PIR > 600) breastfeeding exclusively at 6 months and 46.0% continuing at one year, in comparison with 17.1% and 24.3%, respectively, for households with PIR less than 100. Race/ethnic differences in breastfeeding are also present, although differences across self-identified race/ethnic groups are smaller than those associated with SES (Prevention 2020).

It is important to note that social inequalities in breastfeeding initiation are much smaller than the inequalities in duration: The vast majority of mothers in the US start to breastfeed their babies soon after delivery, but only more educated and affluent women are able to continue breastfeeding for longer periods of time. And even the majority of these women fall short of current AAP recommendations, attesting to the challenges that mothers face in the absence of social policies and cultural norms that support and promote extended breastfeeding in the US (Johnston and Esposito 2007, Hausman 2014).

1.5. Objectives

In the current study, we aim to elucidate the developmental origins of the socioeconomic gradient in chronic inflammation in young adulthood, with a specific focus on LBW and BF duration. We unpack the SES-based gradient in CRP observed in the Add Health cohort with staged structural equation models that introduce both proximate (adult) and distal (early life) contributors to inflammation and its uneven socioeconomic distribution. Our models also highlight the roles of BMI and parent education in understanding these relationships of interest.

2. Methods

2.1. Data

We used data from the National Longitudinal Study of Adolescent to Adult Health (Add Health), a nationally representative cohort of adolescents in the United States in 1994-95 (Harris et al. 2019). The core sample included 20,745 adolescents, 12-19 years old, at Wave 1 (W1). Participants and their parents were recruited with informed consent and interviewed regarding a wide range of social, economic, behavioral, and health measures. For the current study, W1 data were linked to Wave 4 (W4) follow-up data, collected in 2007-8 when participants were 24-32 years old. This dataset included W4 data from in-home interviews and biomarker data (n=15,701) (Whitsel et al. 2012).

2.2. Inflammation

Inflammation was measured using a high sensitivity immunoassay quantifying CRP in dried blood spot (DBS) samples, with values converted to plasma-equivalents (Whitsel et al. 2012) and log-transformed. A single CRP measure provides information on stable, between-individual differences in levels of chronic inflammation that predict future disease risk (Pearson et al. 2003). The DBS samples were collected as part of the W4 in-home interview from 94% of Add Health participants. Among participants that completed both W4 components, 13,166 DBS samples were successfully assayed for CRP. In order to control for acute elevations in inflammation that obscure our ability to determine stable levels of systemic inflammation, we excluded individuals with CRP>10 mg/L, individuals experiencing infectious disease symptoms (cold or flu-like symptoms such as sore throat, runny nose, or cough; fever; night sweats; nausea, vomiting, or diarrhea; bloody stool or urine; frequent urination; or skin rash or abscess) in the two weeks preceding blood collection, and women who reported being pregnant (Pearson et al. 2003, McDade et al. 2014). The remaining sample included 7,610 participants.

2.3. Independent variables

We constructed a SES index informed by multiple indicators established as broadly influential for health. The index included adjusted household income (income divided by the square root of the household size), college completion (categorized as none, some, or completed), tract-level poverty (proportion of population below poverty level in past year), and self-assessed SES (10-point scale) (Ostrove et al. 2000, Braveman et al. 2001). We derived the index from optimally weighted factor scores in Stata using the regression method and a polychoric correlation matrix to account for categorical input variables. Summed scores ranged from 0 to 7, which we then converted to z-scores. As an example of scaling, individuals with z-score values closest to 0 have a mean adjusted household income of $42,885 and 24% likelihood of college completion. In contrast, those with a one-unit greater or lower value have a household income of $73,367 or $20,632 and 57% or 10% likelihood of college completion, respectively (more details online in Table S1).

Birth weight and breastfeeding were measured based on parent interviews at W1. Low birth weight (LBW; ≤ 2,500 grams) was constructed as a dichotomous variable (LBW=1). Breastfeeding duration was constructed as a categorical variable, based on prior research and response options in the W1 survey: none, less than three months, and three months or more (McDade et al. 2014). Body mass in early adulthood was measured as body mass index (BMI), calculated from measured height and weight at W4 and modeled as a continuous variable (kg/m2). Additional covariates included sex, contraceptive use in the past 12 months (reported use of birth control pills, Depo-Provera, Norplant, or an intrauterine device, coil, or loop), anti-inflammatory medication use in the past four weeks (salicylates or nonsteroidal anti-inflammatory, cox-2 inhibitor, inhaled corticosteroid, corticotropin/glucocorticoid, antirheumatic/antipsoriatic, or immunosuppressive medication), smoking regularly in last 30 days (≥ 1 cigarette per day), and parent college completion (graduated from a 4-year college or university, or professional training equivalent).

2.4. Analysis

We first determined the unadjusted SES gradient in chronic inflammation at W4. Based on the distribution of CRP (mg/L), we calculated the gradient as the difference in mean log10CRP associated with a one-standard-deviation difference in SES index, using survey weighting to account for the study design (Chen and Harris 2020). Preliminary analyses also included consideration of bivariate associations between adult SES, parental SES, BMI, birth weight, and breastfeeding.

To assess the relative contributions of LBW and BF to the SES gradient in CRP, we employed generalized linear models of log10CRP in nested stages. First, we adjusted for proximate risk factors not considered to be on the causal pathway linking LBW and BF to CRP (contraceptive use, anti-inflammatory medications, smoking). Next, we added parent education as a potential contributor to both early-life environment and adult inflammation. This model served as our baseline for evaluating the contributions of LBW and BF in the final model set, as indicated by attenuations of the SES gradient with corresponding model stages.

Our final set of models added terms to the baseline model that considered the contributions of LBW and BF, separately and then jointly. This final model set comprised generalized structural equation models (SEM) within which the baseline models were still nested. Each measure was included as a predictor of log10CRP directly and through pathways mediated by BMI (including a quadratic term for nonlinearity). Two-way interaction terms were included between adult SES and both LBW and BF. These interactions allowed for associations to vary by SES, thereby enabling us to assess whether and how improvements to LBW, BF, or both attenuated the SES gradient. To estimate the degree to which specified improvements in these measures would potentially diminish the SES gradient, we estimated predictive margins from each model for comparison. These models produced predicted log10CRP values based on the weighted covariate distributions in the study cohort. For the second stage, or set of models, adjusting for LBW, BF, or both, we focused on conditional predictive margins fixed at advantageous values for breastfeeding duration (≥ three months [BF=3+]) and birth weight (birth weight ≥ 2,500 grams [LBW=0]).

All statistical analyses were performed in Stata 15 (StataCorp, College Station, TX), using gsem, margins, and svy commands. Survey weights were provided by the Add Health study team to account for the study design and missingness (Harris 2013). Characteristics of the analytical subsample were compared with the entire W4 sample with completed interviews, with few differences detected (supplementary Table S2).

We performed multiple sensitivity analyses to assess the robustness of our findings, including replicating our analyses with the inclusion of CRP values above 10 mg/L and respondents reporting recent infectious disease symptoms, alternative specifications of BMI, and adjustments for additional factors potentially underlying inequalities in early life environments and adult inflammation. This included additionally adjusting for maternal education alone, maternal age at birth, birth order, number of siblings, and race and ethnicity.

3. Results

Sample characteristics are summarized in Table 1, including the measures that inform the SES index. Initial bivariate analysis confirmed a strong negative association between adult SES and CRP at ages 24-32 years. This pattern was consistent across transformations of CRP, including a strong linear fit with log10CRP (supplementary Figure S1). Along the fitted line (Figure 1), mean log10CRP equals 0.249 at an SES z-score of −2, or a geometric mean CRP concentration of 1.77 mg/L. At a z-score of 2, mean log10CRP is 0.044, or 1.11 mg/L. The unadjusted SES gradient, or slope, equals a decrease in log10CRP of −0.051 (95% CI: −0.066, −0.036), or a relative difference of 0.889 (0.858, 0.920) in raw CRP with a one-unit higher SES index value.

Table 1.

Descriptive statistics for analytic sample (N=7,610)

Summary
statistic		 Standard
Deviation
Inflammation
Median CRP 1.47
Mean log10CRP 0.15 0.47
Socioeconomic Status
Mean constructed SES index −0.01 1.00
Education
  < High school 9%
  High school 18%
  Some college 43%
  College (4 year) 19%
  More than college 11%
Adjusted household income1
  < $20,000 12%
  $20,000-49,999 32%
  $50,000-74,999 25%
  ≥$75,000 31%
Census track percent population <PL 15%
Mean self-assessed SES (1-10) 5.05 1.72
Parent education
  < High school 16%
  HS equivalent 44%
  Some college 18%
  College (≥4 year) 23%
Early environment
Mean birth weight (grams) 3,366 605
Low birth weight (<2,500 g) 7%
Breastfeeding
  None 55%
  < 3 months 14%
  ≥ 3 months 31%
Other adult risk factors
Mean BMI (kg/m^2) 28.3 6.8
Female sex 43%
Contraceptive use 38%
Regular smoker 25%
Anti-inflammatory medication use 26%
Open in a new tab
1

Household income was estimated based on self-reported, categorical response midpoints and adjusted for household size by dividing by the square root of the total number of household members.

CRP, C-reactive protein

SES, socioeconomic status

BMI, body mass index

Figure 1.

Figure 1.

Open in a new tab

Conditional predictive margins for log10CRP by SES based on nested model stages adjusting for proximate risk factors, parental education, and, finally, adjusting and conditioning on LBW and BF, separately and jointly.

Bivariate analyses confirmed associations between CRP and the early environment (supplementary Figure S2). Individuals who were breastfed for ≥ 3 months had 18% lower CRP. Unadjusted, there was not a clear association between CRP and LBW. Breastfeeding duration and birth weight, along with adult SES, were also patterned by parental education (supplementary Table S3). Having a parent who completed college was associated with a 2.7 times (52% vs. 19%) higher probability of breastfeeding for three or more months, a 6% (181 g) heavier birth weight, a 27% (0.44 mg/L) lower CRP concentration, and a 94% (1.0 standard deviation) higher position on the SES index, relative to having a parent who did not complete high school. We also found that being born LBW was associated with lower odds of breastfeeding for three or more months by 62.4%, adjusting for parent education and maternal prenatal smoking, which further supported our rationale to model LBW and breastfeeding separately and jointly.

Next, we assessed the relative contributions of these variables, along with proximate risk factors, to the observed SES gradient in CRP. The gradient, or slope, estimated from each model using adjusted predictive margins is plotted alongside the unadjusted slope in Figure 1. The model including parent education is the baseline for evaluating how LBW and BF duration account for shifts in the SES gradient in CRP.

Results indicate a strong attenuation of the SES gradient in CRP when BF, LBW, and BMI are added to the models. This attenuation is summarized in Table 2. Relative to the baseline, accounting for LBW and setting LBW at 0 was associated with a 54% reduction in the slope associating SES and CRP. Breastfeeding plays an even stronger role: When BF duration is 3 months or greater, the SES gradient in CRP is reduced 80%. Adjusting for both LBW and BF was associated with an 83% reduction.

Table 2.

SES gradient in log10CRP estimated using conditional predictive marginal differences with four different models and specifications for LBW and BF. The baseline (top left) adjusted for parent education, sex, contraceptive use, smoking, and anti-inflammatory medication.

Without LBW LBW- and BMI-adjusted &
conditioned on LBW=0
Without BF −0.036
(−0.053, −0.018)		 −0.017
(−0.031, −0.002)
Relative (raw) difference 0.921 0.962
Percent reduction 54%
BF- and BMI-adjusted & conditioned on BF= 3+ months −0.007
(−0.033, 0.019)		 −0.006
(−0.032, 0.019)
Relative (raw) difference 0.984 0.986
Percent reduction 80% 83%
Open in a new tab

Using the same fully adjusted SEM above, we examined the patterns of association among key variables, with specific attention to BMI as a potential mediator linking LBW and BF to adult CRP (Figure 2). To aid the interpretation of non-linear terms, measures of association are presented as predictive marginal differences (not coefficients), i.e., derivatives of adjusted mean predicted CRP based on each cohort member’s predictor values. Predictive marginal differences offer summary measures of associations of interest that are standardized by cohort member characteristics to incorporate interactions and polynomial terms. These are listed both on the log scale and transformed to represent relative differences in raw CRP. For example, the numerical derivative of log10CRP with respect to BMI is 0.04. This represents the adjusted difference in log10CRP associated with a 1 kg/m2 difference in BMI averaged across individual adjusted predicted values, which equates to a 9% increase in raw CRP. Breastfeeding for three or more months is associated with a lower BMI by 0.63 kg/m2, relative to not breastfeeding; and LBW is associated with a lower BMI by 1.02 kg/m2. The summary estimates linking LBW and BF directly to CRP are calculated similarly, as marginal differences adjusted for BMI. All SEM coefficients are listed in supplementary Table S4.

Figure 2.

Figure 2.

Open in a new tab

Summary of full SEM results using predictive marginal differences with the Add Health cohort at W4, survey weighted. The summary measures referencing categorical variables compare the highest and lowest value groups. Each arrow value represents marginal differences (not coefficients) that account for interactions and polynomial terms. Differences in log(CRP) are transformed to reflect relative differences in parentheses. Statistical test results indicated as: *p<0.05, **p<0.01, ***p<0.001.

Overall, SEM results indicate that links between LBW, BF duration, and CRP are largely mediated through their associations with BMI, which was positively associated with CRP in both unadjusted (supplementary Figure S3) and adjusted analyses (supplementary Table S4). Comparisons with models omitting BMI (supplementary Table S4) reveal that BMI accounts for 40% of the marginal difference in log10CRP associated with breastfeeding ≥ 3 months. The direct association between LBW and higher CRP is only revealed after taking into account BMI as a mediator, which is explained jointly by the negative association between LBW and BMI and the positive association between BMI and CRP. Additionally, when BMI is not taken into account LBW and BF ≥ 3 months are associated with smaller reductions in the SES gradient in CRP (supplementary Table S6).

Finally, as expected, parent education is additionally important for adult inflammation in multiple ways. Beyond patterning birth weight, breastfeeding, adult SES, and BMI, it also maintains an independent association with adult CRP. After adjusting for other independent and mediating variables in our model, individuals with a parent who completed college still experience a −0.04 (95% CI: −0.07, −0.00) lower log10CRP than individuals with a parent who did not complete high school. This equates to a reduction in CRP of 0.92 mg/L (0.84, 1.00).

Sensitivity analyses suggest that findings and their interpretation are robust to modeling choices related to our exclusion criteria and additional adjustments (Table S6). Replicating our analyses with the inclusion of CRP values above 10 mg/L resulted in a steeper SES slope in log10CRP, with more of the slope accounted for by LBW and BF duration. Associations were attenuated when individuals with infectious symptoms at the time of blood collection were included. Additional baseline adjustments for maternal education, age at birth, number of siblings, birth order, and race and ethnicity all resulted in SES slopes and attenuation similar to the main analysis. Adjusting for BMI at baseline led to smaller estimates of the slope and attenuation, as expected. Not adjusting for BMI at all led to the least amount of attenuation across adjustments, as discussed above.

4. Discussion

Social inequalities in cardiometabolic diseases in the US are large and growing. Environments early in life shape the development of physiological systems and have lasting effects on health in adulthood. In this analysis we weave these threads together to investigate the extent to which the SES gradient in chronic inflammation in adulthood can be explained by patterns of birth weight and breastfeeding in infancy. Results indicate that environments early in infancy are important determinants of chronic inflammation in adulthood, and that breastfeeding in particular plays an important role in shaping the SES gradient in CRP.

First, our study documents substantial inequalities in chronic inflammation in a large, representative cohort of young adults in the US, with lower SES associated with significantly higher concentrations of CRP. This observation is consistent with prior studies of older adults (Alley et al. 2006, Nazmi and Victora 2007, Ranjit et al. 2007, McDade et al. 2010), but the magnitude of the SES gradient in CRP is particularly striking given that the sample is comprised of young adults. CRP is a strong, preclinical predictor of cardiometabolic disease and overall health, and it is clear that processes contributing to health inequalities are in motion in young adulthood despite the fact that the prevalence of clinical disease is relatively low compared with later stages of the life course.

Second, we uncover important contributions of the environment in infancy to the SES gradient in CRP in adulthood. In particular, our models point toward breastfeeding as playing a central role in the developmental origins of social inequities in chronic inflammation. Members of the Add Health cohort were born in the late 1970s and early 1980s, during a time when rates of breastfeeding in the US were on the rise after falling to historic lows (Medicine et al. 1991, Ryan 1997). In our sample, more than half the participants were not breastfed at all, and less than one in three was breastfed for three months or longer. However, if all of the members of the cohort had been breastfed for ≥3 months, our estimates suggest that the SES gradient in chronic inflammation in young adulthood would be reduced by 80%. By contrast, low birth weight—a widely studied marker of the prenatal environment—accounts for only 54% of the SES gradient in CRP.

While LBW is associated with shorter breastfeeding in this cohort, our results across models suggest that breastfeeding’s contribution is independent of LBW. At the same time, more work is needed to further disentangle the contributions of LBW, as it may have cascading impacts that counteract each other, e.g., simultaneously suppressing breastfeeding and BMI while elevating chronic inflammation through direct effects on immune development and regulation. It is possible that these pathways may be more or less influential depending on other contextual factors, which may explain the negative or neutral impact of LBW on the SES-based gradient in adult CRP despite the negative LBW-BMI association and the positive BMI-CRP association we observe.

It is also important to keep in mind that our estimates are based on observational data, and while we control for a wide range of potentially confounding influences it is possible that the associations among breastfeeding, birthweight, BMI, and CRP are not causal. There may be unmeasured confounders, including genetic factors, that influence propensity of breastfeeding, LBW, and chronic inflammation. Regarding the breastfeeding-inflammation association, we re-ran our analyses with birthweight modeled continuously in order to more precisely adjust for any factors that would affect body mass at birth along with breastfeeding, adult BMI, and inflammation. Our results did not change. Ideally, we would adjust for parent height, weight, or both to better account for this, especially in the case of LBW, but these data were unavailable. Secondly, we cannot entirely rule out the possibility that inflammation also influences BMI. If this were the case, it could bias our full model estimates when we condition on BMI (Greenland 2003). However, we considered multiple scenarios of how this would occur using a directed acyclic graph and deemed it unlikely based on our model results (Figure S4).

Our findings build upon, and are consistent with, prior studies that suggest significant long term effects of breastfeeding and birth weight with quasi-experimental research designs (Metzger and McDade 2010, McDade et al. 2014). Of note, the magnitude of association between breastfeeding duration and BMI in our analysis is large in comparison with other cohorts and systematic reviews (Gillman et al. 2006, Horta et al. 2015). The reasons are not clear, but may reflect in part how we categorize breastfeeding duration, and the fact that this cohort of younger adults has come of age during a pronounced secular increase in BMI in the US (Hales et al. 2017).

Furthermore, biological mechanisms through which breastfeeding can exert lasting effects on immune development are well-established (Field 2005, Cacho and Lawrence 2017, Vieira Borba et al. 2018). Breastmilk is comprised of bioactive proteins and cells that not only protect against infection, but also promote the maturation and regulation of the infant’s immune defenses. In addition, human milk oligosaccharides serve as prebiotics that, along with the microbiota in milk, promote the establishment of a healthy intestinal microbiota in infancy, which in turn modulates intestinal barrier function and regulates local and systemic inflammatory responses (Stiemsma and Michels 2018). Lastly, breastfeeding can also have indirect effects on inflammation through the programming of metabolic pathways that contribute to the development of overweight/obesity and the production of pro-inflammatory cytokines in adipose tissue (Harder et al. 2005, Schaffler et al. 2006). Our results underscore the potential importance of this latter pathway and suggest that future research should consider whether patterns of association among breastfeeding, overweight/obesity, and chronic inflammation are similar in different cohorts and epidemiological contexts.

This study makes an important contribution in evaluating the developmental origins of social inequities in health in adulthood. Using a consequentialist epidemiological approach, we go beyond the focus on individual risk factors to illuminate the processes that lead to inequities in population-based risk (Keyes and Galea 2015). While prior research has documented how exposures in infancy shape individual trajectories of later life health, we consider the implications for population health by demonstrating how shifts in breastfeeding and birthweight can modify distributions of chronic inflammation in adulthood. By focusing on the SES gradient in inflammation as our primary outcome, we prioritize actionable goals that would have the greatest impact on reducing social inequities in health.

In particular, our models suggest that extending the duration of breastfeeding to three months or longer—a modest goal in relation to current AAP guidelines—could dramatically shift the population distribution of chronic inflammation and reduce SES-based inequities in cardiometabolic disease risk. Barriers to extended breastfeeding in the US are substantial, but there is clear evidence for policies that effectively promote the initiation and continuation of breastfeeding (Baker and Milligan 2008, Pérez-Escamilla et al. 2016, Hamad et al. 2019). While many programs focus on attitudes and knowledge of mothers, fathers, and healthcare providers, social policies that address structural barriers to extended breastfeeding (e.g., paid family leave) will likely have greater impact on reducing social inequities in breastfeeding.

Our study additionally highlights the importance of parental SES and the role of intergenerational social advantage and disadvantage that continue to shape population health inequities. In our analyses, parental education predicts birthweight and breastfeeding, but remains independently and significantly associated with lower CRP. Any successful intervention on early life environments must jointly address socioeconomic barriers faced by parents of young children in ways that structurally break intergenerational chains of poverty and disadvantage.

Our conclusions are based the analysis of data from a large, nationally representative cohort of young adults. However, the study is observational and while we apply an adjusted structural modeling approach with multiple robustness tests, we cannot completely rule out the possibility that omitted variables may bias associations of interest. Additional limitations include our use of a single measure of CRP to index chronic inflammation, and our reliance on maternal recall to provide information on birth weight and breastfeeding duration. However, maternal reports of birth weight correlate highly with birth records, and recalls of breastfeeding initiation and duration have been shown to be reported with high validity and reliability (Tomeo et al. 1999, Li et al. 2008). Lastly, the Add Health study did not collect information on gestational age, so it is not possible for us to determine whether associations between LBW and CRP are due to pre-term vs. small-for-gestational age deliveries.

This study contributes to existing research that has revealed how early-life development and nutrition affect health over the life course by examining more closely how these processes underlie socioeconomic inequities in U.S. adult health. In doing so, it builds beyond an understanding of development as a risk factor for adult health and toward a more consequential epidemiological understanding of its critical role in eliminating persistent health inequities affecting the U.S. population.

Supplementary Material

1

Highlights.

Lower socioeconomic status is associated with higher chronic inflammation

Breastfed individuals have lower levels of chronic inflammation in young adulthood

Breastfeeding accounts for 80 percent of the socioeconomic gradient in inflammation

Body mass partly mediates the association between breastfeeding and inflammation

Acknowledgements

Research reported in this publication was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health under Award Number R21HD101757. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. This research uses data from Add Health, a program project directed by Kathleen Mullan Harris and designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris at the University of North Carolina at Chapel Hill, and funded by grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 23 other federal agencies and foundations. Information on how to obtain the Add Health data files is available on the Add Health website (https://dev-addhealth.cpc.unc.edu/). No direct support was received from grant P01-HD31921 for this analysis.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Adler NE, Boyce T, Chesney MA, Cohen S, Folkman S, Kahn RL and Syme SL (1994). "Socioeconomic status and health: The challenge of the gradient." American Psychologist 49(1): 15–24. [DOI] [PubMed] [Google Scholar]
  2. Adler NE and Snibbe AC (2003). "The role of psychosocial processes in explaining the gradient between socioeconomic status and health." Current Directions in Psychological Science 12(4): 119–123. [Google Scholar]
  3. Aiello AE and Kaplan GA (2009). "Socioeconomic position and inflammatory and immune biomarkers of cardiovascular disease: applications to the Panel Study of Income Dynamics." Biodemography Soc Biol 55(2): 178–205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Alley DE, Seeman TE, Ki Kim J, Karlamangla A, Hu P and Crimmins EM (2006). "Socioeconomic status and C-reactive protein levels in the US population: NHANES IV." Brain, Behavior, and Immunity 20(5): 498–504. [DOI] [PubMed] [Google Scholar]
  5. Alley DE, Seeman TE, Ki Kim J, Karlamangla A, Hu P and Crimmins EM (2006). "Socioeconomic status and C-reactive protein levels in the US population: NHANES IV." Brain, Behavior, & Immunity 20(5): 498–504. [DOI] [PubMed] [Google Scholar]
  6. Baker M and Milligan K (2008). "Maternal employment, breastfeeding, and health: Evidence from maternity leave mandates." Journal of Health Economics 27(4): 871–887. [DOI] [PubMed] [Google Scholar]
  7. Barker DJ (1998). "In utero programming of chronic disease." Clinical Science 95: 115–128. [PubMed] [Google Scholar]
  8. Barker DJ, Bagby SP and Hanson MA (2006). "Mechanisms of disease: in utero programming in the pathogenesis of hypertension." Nature Clinical Practice Nephrology 2(12): 700–707. [DOI] [PubMed] [Google Scholar]
  9. Barker DJ, Eriksson JG, Forsen T and Osmond C (2002). "Fetal origins of adult disease: strength of effects and biological basis." Int J Epidemiol 31(6): 1235–1239. [DOI] [PubMed] [Google Scholar]
  10. Barker DJ, Osmond C, Golding J, Kuh D and Wadsworth M (1989). "Growth in utero, blood pressure in childhood and adult life, and mortality from cardiovascular disease." British Medical Journal 298(6673): 564–567. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Bengtsson T and Broström G (2009). "Do conditions in early life affect old-age mortality directly and indirectly? Evidence from 19th-century rural Sweden." Social Science & Medicine 68(9): 1583–1590. [DOI] [PubMed] [Google Scholar]
  12. Berkman LF, Kawachi I and Glymour MM (2014). Social Epidemiology, Oxford University Press. [Google Scholar]
  13. Black S, Kushner I and Samols D (2004). "C-reactive Protein." Journal of Biological Chemistry 279(47): 48487–48490. [DOI] [PubMed] [Google Scholar]
  14. Blackwell DL, Hayward MD and Crimmins EM (2001). "Does childhood health affect chronic morbidity in later life?" Social Science & Medicine 52(8): 1269–1284. [DOI] [PubMed] [Google Scholar]
  15. Blumenshine P, Egerter S, Barclay CJ, Cubbin C and Braveman PA (2010). "Socioeconomic disparities in adverse birth outcomes: a systematic review." American Journal of Preventive Medicine 39(3): 263–272. [DOI] [PubMed] [Google Scholar]
  16. Braveman P, Cubbin C, Marchi K, Egerter S and Chavez G (2001). "Measuring socioeconomic status/position in studies of racial/ethnic disparities: maternal and infant health." Public Health Reports 116(5): 449. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Breastfeeding, S. (2012). "Breastfeeding and the use of human milk." Pediatrics 129(3): e827–e841. [DOI] [PubMed] [Google Scholar]
  18. Cacho NT and Lawrence RM (2017). "Innate immunity and breast milk." Frontiers in Immunology 8: 584. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Cameron N and Demerath EW (2002). "Critical periods in human growth and their relationship to diseases of aging." Yearbook of Physical Anthropology, Vol 45 45: 159–184. [DOI] [PubMed] [Google Scholar]
  20. Case A and Paxson C (2010). "Causes and consequences of early-life health." Demography 47(1): S65–S85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Catov JM, Lee M, Roberts JM, Xu J and Simhan HN (2015). "Race disparities and decreasing birth weight: are all babies getting smaller?" American Journal of Epidemiology 183(1): 15–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Challis JR, Lockwood CJ, Myatt L, Norman JE, Strauss JF and Petraglia F (2009). "Inflammation and Pregnancy." Reproductive Sciences 16(2): 206–215. [DOI] [PubMed] [Google Scholar]
  23. Chen P and Harris KM (2020). "Guidelines for analyzing Add Health data." https://addhealth.cpc.unc.edu/wp-content/uploads/docs/user_guides/GuidelinesforAnalysisofAddHealthData_202004.pdf. [Google Scholar]
  24. Cohen DA, Farley TA and Mason K (2003). "Why is poverty unhealthy? Social and physical mediators." Social Science & Medicine 57: 1631–1641. [DOI] [PubMed] [Google Scholar]
  25. Cole SW, Hawkley LC, Arevalo JM, Sung CY, Rose RM and Cacioppo JT (2007). "Social regulation of gene expression: Inflammation and human transcriptional response to loneliness." Genome Biol 8(9): R189. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Crimmins EM and Finch CE (2006). "Infection, inflammation, height, and longevity." Proc Natl Acad Sci U S A 103(2): 498–503. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Danese A, Pariante CM, Caspi A, Taylor A and Poulton R (2007). "Childhood maltreatment predicts adult inflammation in a life-course study." Proc Natl Acad Sci U S A 104(4): 1319–1324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Field CJ (2005). "The immunological components of human milk and their effect on immune development in infants." The Journal of nutrition 135(1): 1–4. [DOI] [PubMed] [Google Scholar]
  29. Forbes JD, Azad MB, Vehling L, Tun HM, Konya TB, Guttman DS, Field CJ, Lefebvre D, Sears MR and Becker AB (2018). "Association of exposure to formula in the hospital and subsequent infant feeding practices with gut microbiota and risk of overweight in the first year of life." JAMA pediatrics 172(7): e181161–e181161. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Ford ES, Giles WH, Mokdad AH and Myers GL (2004). "Distribution and correlates of C-reactive protein concentrations among adult US women." Clinical Chemistry 50(3): 574–581. [DOI] [PubMed] [Google Scholar]
  31. Forsdahl A (1977). "Are Poor Living conditions in Childhood and Adolescence an Important Risk Factor for Arteriosclerotic Heart Disease?" British Journal of Preventive and Social Medicine 31: 91–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Gillman MW, Rifas-Shiman SL, Berkey CS, Frazier AL, Rockett HR, Camargo CA Jr, Field AE and Colditz GA (2006). "Breast-feeding and overweight in adolescence: within-family analysis." Epidemiology 17(1): 112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Gluckman PD, Buklijas T and Hanson MA (2016). The developmental origins of health and disease (DOHaD) concept: past, present, and future The epigenome and developmental origins of health and disease, Elsevier: 1–15. [DOI] [PubMed] [Google Scholar]
  34. Gluckman PD, Hanson MA, Cooper C and Thornburg KL (2008). "Effect of in utero and early-life conditions on adult health and disease." New England Journal of Medicine 359(1): 61–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Goosby BJ, Cheadle JE and McDade T (2016). "Birth weight, early life course BMI, and body size change: Chains of risk to adult inflammation?" Social Science & Medicine 148: 102–109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Goosby BJ, Cheadle JE and Mitchell C (2018). "Stress-related biosocial mechanisms of discrimination and African American health inequities." Annual Review of Sociology 44: 319–340. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Gravlee CC (2009). "How race becomes biology: Embodiment of social inequality." American Journal of Physical Anthropology 139(1): 47–57. [DOI] [PubMed] [Google Scholar]
  38. Green T and Hamilton TG (2019). "Maternal educational attainment and infant mortality in the United States." Demographic Research 41: 713–752. [Google Scholar]
  39. Greenland S (2003). "Quantifying biases in causal models: classical confounding vs collider-stratification bias." Epidemiology 14(3): 300–306. [PubMed] [Google Scholar]
  40. Hales CM, Carroll MD, Fryar CD and Ogden CL (2017). "Prevalence of obesity among adults and youth: United States, 2015–2016." [PubMed] [Google Scholar]
  41. Halfon N, Larson K, Lu M, Tullis E and Russ S (2014). "Lifecourse health development: past, present and future." Maternal and child health journal 18(2): 344–365. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Hamad R, Modrek S and White JS (2019). "Paid family leave effects on breastfeeding: a quasi-experimental study of US policies." American Journal of Public Health 109(1): 164–166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Harder T, Bergmann R, Kallischnigg G and Plagemann A (2005). "Duration of breastfeeding and risk of overweight: a meta-analysis." American Journal of Epidemiology 162(5): 397–403. [DOI] [PubMed] [Google Scholar]
  44. Harris KM (2013). "The Add Health study: Design and accomplishments." https://addhealth.cpc.unc.edu/wp-content/uploads/docs/user_guides/DesignPaperWave_I-IV.pdf. [Google Scholar]
  45. Harris KM, Halpern CT, Whitsel EA, Hussey JM, Killeya-Jones LA, Tabor J and Dean SC (2019). "Cohort profile: The national longitudinal study of adolescent to adult health (Add health)." International Journal of Epidemiology 48(5): 1415–1415k. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Harris TB, Ferrucci L, Tracy RP, Corti MC, Wacholder S, Ettinger WH, Heimovitz H, Cohen HJ and Wallace R (1999). "Associations of elevated interleukin-6 and C-reactive protein levels with mortality in the elderly." American Journal of Medicine 106: 516–512. [DOI] [PubMed] [Google Scholar]
  47. Hausman BL (2014). Mother's milk: Breastfeeding controversies in American culture, Routledge. [Google Scholar]
  48. Hayward MD and Gorman BK (2004). "The long arm of childhood: The influence of early-life social conditions on men's mortality." Demography 41(1): 87–107. [DOI] [PubMed] [Google Scholar]
  49. Hertzman C and Boyce T (2010). "How experience gets under the skin to create gradients in developmental health." Annu Rev Public Health 31: 329–347 323p following 347. [DOI] [PubMed] [Google Scholar]
  50. Hoke MK and McDade T (2014). "Biosocial inheritance: A framework for the study of the intergenerational transmission of health disparities." Annals of Anthropological Practice 38(2): 187–213. [Google Scholar]
  51. Horta BL, Loret de Mola C and Victora CG (2015). "Long-term consequences of breastfeeding on cholesterol, obesity, systolic blood pressure and type 2 diabetes: a systematic review and meta-analysis." Acta Paediatrica 104: 30–37. [DOI] [PubMed] [Google Scholar]
  52. Jackson JS, Knight KM and Rafferty JA (2010). "Race and unhealthy behaviors: chronic stress, the HPA axis, and physical and mental health disparities over the life course." American Journal of Public Health 100(5): 933–939. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Jenny NS, Yanez ND, Psaty BM, Kuller LH, Hirsch CH and Tracy RP (2007). "Inflammation biomarkers and near-term death in older men." American Journal of Epidemiology 165(6): 684–695. [DOI] [PubMed] [Google Scholar]
  54. Johnston ML and Esposito N (2007). "Barriers and facilitators for breastfeeding among working women in the United States." Journal of Obstetric, Gynecologic & Neonatal Nursing 36(1): 9–20. [DOI] [PubMed] [Google Scholar]
  55. Keppel KG, Pearcy JN and Wagener DK (2002). "Trends in racial and ethnic-specific rates for the health status indicators: United States, 1990-98." Healthy People 2000 Stat Notes(23): 1–16. [PubMed] [Google Scholar]
  56. Kermack WO, McKendrick AG and McKinlay PL (1934). "Death Rates in Great Britain and Sweden: Some General Regularities and Their Significance." Lancet 226: 698–703. [DOI] [PubMed] [Google Scholar]
  57. Keyes K and Galea S (2015). "What matters most: quantifying an epidemiology of consequence." Annals of Epidemiology 25(5): 305–311. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Kittleson MM, Meoni LA, Wang NY, Chu AY, Ford DE and Klag MJ (2006). "Association of childhood socioeconomic status with subsequent coronary heart disease in physicians." Arch Intern Med 166(21): 2356–2361. [DOI] [PubMed] [Google Scholar]
  59. Kuo H-K, Bean JF, Yen C-J and Leveille SG (2006). "Linking C-reactive protein to late-life disability in the National Health and Nutrition Examination Survey (NHANES) 1999-2002." The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 61(4): 380–387. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Kuo H-K, Yen C-J, Chang C-H, Kuo C-K, Chen J-H and Sorond F (2005). "Relation of C-reactive protein to stroke, cognitive disorders, and depression in the general population: systematic review and meta-analysis." The Lancet Neurology 4(6): 371–380. [DOI] [PubMed] [Google Scholar]
  61. Kuzawa CW and Sweet E (2009). "Epigenetics and the embodiment of race: developmental origins of US racial disparities in cardiovascular health." Am J Hum Biol 21(1): 2–15. [DOI] [PubMed] [Google Scholar]
  62. Li R, Scanlon KS and Serdula MK (2008). "The validity and reliability of maternal recall of breastfeeding practice." Nutrition reviews 63(4): 103–110. [DOI] [PubMed] [Google Scholar]
  63. Mackenbach JP and Kunst AE (1997). "Measuring the magnitude of socio-economic inequalities in health: An overview of available measures illustrated with two examples from Europe." Social Science & Medicine 44(6): 757–771. [DOI] [PubMed] [Google Scholar]
  64. Marmot M, Ryff CD, Bumpass LL, Shipley M and Marks NF (1997). "Social inequalities in health: Next questions and converging evidence." Social Science & Medicine 44(6): 901–910. [DOI] [PubMed] [Google Scholar]
  65. Martin J, Hamilton B, Osterman M, Curtin S and Matthews T (2015). "Births: final data for 2013." National vital statistics reports: from the Centers for Disease Control and Prevention, National Center for Health Statistics, National Vital Statistics System 64(1): 1. [PubMed] [Google Scholar]
  66. Mathews T and MacDorman MF (2013). "Infant mortality statistics from the 2010 period linked birth/infant death data set." Natl Vital Stat Rep 62(8): 1–26. [PubMed] [Google Scholar]
  67. McDade TW (2006). "Life history, maintenance, and the early origins of immune function. ." American Journal of Human Biology 17: 81–94. [DOI] [PubMed] [Google Scholar]
  68. McDade TW (2012). "Early environments and the ecology of inflammation." Proceedings of the National Academy of Sciences 109: 17281–17288. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. McDade TW, Hawkley LC and Cacioppo JT (2006). "Psychosocial and behavioral predictors of inflammation in middle-aged and older adults: the Chicago health, aging, and social relations study." Psychosom Med 68(3): 376–381. [DOI] [PubMed] [Google Scholar]
  70. McDade TW, Lindau ST and Wroblewski K (2010). "Predictors of C-Reactive Protein in the National Social Life, Health, and Aging Project." J Gerontol B Psychol Sci Soc Sci. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. McDade TW, Metzger MW, Chyu L, Duncan GJ, Garfield C and Adam EK (2014). "Long-term effects of birth weight and breastfeeding duration on inflammation in early adulthood." Proceedings of the Royal Society B-Biological Sciences 281(1784). [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. McDade TW, Rutherford J, Adair L and Kuzawa CW (2010). "Early origins of inflammation: microbial exposures in infancy predict lower levels of C-reactive protein in adulthood." Proceedings of the Royal Society B 277: 1129–1137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Medicine, I. o., N. R. C. S. o. N. D. Lactation, I. o. M. S. o. N. d. Lactation, I. o. M. S. o. Lactation, U. S. H. Resources, S. Administration, I. o. M. Staff, É.-U. H. Resources, S. Administration and C. o. N. S. D. Pregnancy (1991). Nutrition during lactation, Henry Holt and Company. [Google Scholar]
  74. Metzger MW and McDade TW (2010). "Breastfeeding as obesity prevention in the United States: a sibling difference model." Am J Hum Biol 22(3): 291–296. [DOI] [PubMed] [Google Scholar]
  75. Miller GE, Chen E, Fok AK, Walker H, Lim A, Nicholls EF, Cole S and Kobor MS (2009). "Low early-life social class leaves a biological residue manifested by decreased glucocorticoid and increased proinflammatory signaling." Proc Natl Acad Sci U S A 106(34): 14716–14721. [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Miller GE, Cohen S and Ritchey AK (2002). "Chronic psychological stress and the regulation of pro-inflammatory cytokines: a glucocorticoid-resistance model." Health Psychol 21(6): 531–541. [DOI] [PubMed] [Google Scholar]
  77. Miller GE, Freedland KE, Carney RM, Stetler CA and Banks WA (2003). "Pathways linking depression, adiposity, and inflammatory markers in healthy young adults." Brain Behav Immun 17(4): 276–285. [DOI] [PubMed] [Google Scholar]
  78. Morisaki N, Esplin MS, Varner MW, Henry E and Oken E (2013). "Declines in birth weight and fetal growth independent of gestational length." Obstetrics and Gynecology 121(1): 51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  79. Nazmi A and Victora CG (2007). "Socioeconomic and racial/ethnic differentials of C-reactive protein levels: a systematic review of population-based studies." BMC Public Health 7(212): doi: 10.1186/1471-2458-1187-1212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  80. Nazmi A and Victora CG (2007). "Socioeconomic and racial/ethnic differentials of C-reactive protein levels: a systematic review of population-based studies." BMC Public Health 7: 212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  81. Olson JS and Hayward MD (2017). "Breastfeeding, overweight status, and inflammation." Social Science Research 64: 226–236. [DOI] [PMC free article] [PubMed] [Google Scholar]
  82. Ostrove JM, Adler NE, Kuppermann M and Washington AE (2000). "Objective and subjective assessments of socioeconomic status and their relationship to self-rated health in an ethnically diverse sample of pregnant women." Health Psychology 19(6): 613. [DOI] [PubMed] [Google Scholar]
  83. Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon RO, Criqui M, Fadl YY, Fortmann SP, Hong Y, Myers GL, Rifai N, Smith SC, Taubert K, Tracy RP and Vinicor F (2003). "Markers of inflammation and cardiovascular disease: Application to clinical and public health practice." Circulation 107: 499–511. [DOI] [PubMed] [Google Scholar]
  84. Pérez-Escamilla R, Martinez JL and Segura-Pérez S (2016). "Impact of the Baby-friendly Hospital Initiative on breastfeeding and child health outcomes: a systematic review." Maternal & Child Nutrition 12(3): 402–417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  85. Pradhan AD, Manson JE, Rifai N, Buring JE and Ridker PM (2001). "C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus." Journal of American Medical Association 286(3): 327–334. [DOI] [PubMed] [Google Scholar]
  86. Preston SH, Hill ME and Drevenstedt GL (1998). "Childhood conditions that predict survival to advanced ages among African–Americans." Social Science & Medicine 47(9): 1231–1246. [DOI] [PubMed] [Google Scholar]
  87. Prevention, C. f. D. C. a. (2020). "Breastfeeding among U.S. children born 2010-2017, CDC National Immunization Survey." Retrieved June 15, 2020, 2020, from https://www.cdc.gov/breastfeeding/data/nis_data/results.html.
  88. Ranjit N, Diez-Roux AV, Shea S, Cushman M, Hanyu N and Seeman T (2007). "Socioeconomic Position, Race/Ethnicity, and Inflammation in the Multi-Ethnic Study of Atherosclerosis. ." Circulation 116: 2383–2390. . [DOI] [PubMed] [Google Scholar]
  89. Ranjit N, Diez-Roux AV, Shea S, Cushman M, Ni H and Seeman T (2007). "Socioeconomic position, race/ethnicity, and inflammation in the multi-ethnic study of atherosclerosis." Circulation 116(21): 2383–2390. [DOI] [PubMed] [Google Scholar]
  90. Ranjit N, Diez-Roux AV, Shea S, Cushman M, Seeman T, Jackson SA and Ni H (2007). "Psychosocial factors and inflammation in the multi-ethnic study of atherosclerosis." Arch Intern Med 167(2): 174–181. [DOI] [PubMed] [Google Scholar]
  91. Ridker PM, Buring JE, Cook NR and Rifai N (2003). "C-reactive protein, the metabolic syndrome, and risk of incident cardiovascular events: An 8-Year follow-up of 14 719 initially healthy American women." Circulation 107(3): 391–397. [DOI] [PubMed] [Google Scholar]
  92. Romero R, Gotsch F, Pineles B and Kusanovic JP (2007). "Inflammation in pregnancy: Its roles in reproductive physiology, obstetrical complications, and fetal injury." Nutrition Reviews 65: S194–S202. [DOI] [PubMed] [Google Scholar]
  93. Rudnicka AR, Owen CG and Strachan DP (2007). "The effect of breastfeeding on cardiorespiratory risk factors in adult life." Pediatrics 119(5): e1107–e1115. [DOI] [PubMed] [Google Scholar]
  94. Ryan AS (1997). "The resurgence of breastfeeding in the United States." Pediatrics 99(4): e12–e12. [DOI] [PubMed] [Google Scholar]
  95. Sattar N, McConnachie A, O'Reilly D, Upton MN, Greer IA, Smith GD and Watt G (2004). "Inverse association between birth weight and C-reactive protein concentrations in the MIDSPAN Family Study." Arteriosclerosis, Thrombosis, and Vascular Biology 24(3): 583–587. [DOI] [PubMed] [Google Scholar]
  96. Schaffler A, Muller-Ladner U, Scholmerich J and Buchler C (2006). "Role of adipose tissue as an inflammatory organ in human diseases." Endocrine Reviews 27(5): 449–467. [DOI] [PubMed] [Google Scholar]
  97. Shanks N and Lightman SL (2001). "The maternal-neonatal neuro-immune interface: are there long-term implications for inflammatory or stress-related disease?" J Clin Invest 108(11): 1567–1573. [DOI] [PMC free article] [PubMed] [Google Scholar]
  98. Shonkoff JP, Boyce WT and McEwen BS (2009). "Neuroscience, molecular biology, and the childhood roots of health disparities: building a new framework for health promotion and disease prevention." JAMA 301(21): 2252–2259. [DOI] [PubMed] [Google Scholar]
  99. Smith CJ and Ryckman KK (2015). "Epigenetic and developmental influences on the risk of obesity, diabetes, and metabolic syndrome." Diabetes Metab Syndr Obes 8: 295–302. [DOI] [PMC free article] [PubMed] [Google Scholar]
  100. Statistics, N. C. f. H. (2012). "Health, United States, 2011: With special feature on socioeconomic status and health." [PubMed] [Google Scholar]
  101. Stiemsma LT and Michels KB (2018). "The role of the microbiome in the developmental origins of health and disease." Pediatrics 141(4). [DOI] [PMC free article] [PubMed] [Google Scholar]
  102. Taylor SE (2010). "Mechanisms linking early life stress to adult health outcomes." Proc Natl Acad Sci U S A 107(19): 8507–8512. [DOI] [PMC free article] [PubMed] [Google Scholar]
  103. Taylor SE, Lehman BJ, Kiefe CI and Seeman TE (2006). "Relationship of early life stress and psychological functioning to adult C-reactive protein in the coronary artery risk development in young adults study." Biol Psychiatry 60(8): 819–824. [DOI] [PubMed] [Google Scholar]
  104. Thayer ZM and Kuzawa CW (2013). "Intergenerational effects of maternal experience: Influence of SES on maternal and offspring stress physiology in New Zealand." American Journal of Human Biology 25(2): 277–277. [Google Scholar]
  105. Tomeo CA, Rich-Edwards JW, Michels KB, Berkey CS, Hunter DJ, Frazier AL, Willett WC and Buka SL (1999). "Reproducibility and validity of maternal recall of pregnancy-related events." Epidemiology: 774–777. [PubMed] [Google Scholar]
  106. Tzoulaki I, Jarvelin MR, Hartikainen AL, Leinonen M, Pouta A, Paldanius M, Ruokonen A, Canoy D, Sovio U, Saikku P and Elliott P (2008). "Size at birth, weight gain over the life course, and low-grade inflammation in young adulthood: northern Finland 1966 Birth Cohort study." Eur Heart J 29(8): 1049–1056. [DOI] [PubMed] [Google Scholar]
  107. Vieira Borba V, Sharif K and Shoenfeld Y (2018). "Breastfeeding and autoimmunity: Programing health from the beginning." American Journal of Reproductive Immunology 79(1): e12778. [DOI] [PubMed] [Google Scholar]
  108. Whitsel E, Cuthbertson CC, Tabor JW, Potter AJ, Wener MH, Killeya-Jones LA and Harris KM (2012). Add Health Wave IV Documentation: Measures of Inflammation and Immune Function. Chapel Hill, NC, Carolina Population Center, Univerisyt of North Carolina at Chapel Hill. [Google Scholar]
  109. Whitsel EA, Cuthbertson CC, Tabor JW, Potter AJ, Wener MH, Killeya-Jones LA and Harris KM (2012). "Add Health wave IV documentation: Measures of inflammation and immune function." [Google Scholar]
  110. Williams M, Williams SM and Poulton R (2006). "Breast feeding is related to C reactive protein concentration in adult women." Journal of Epidemiology & Community Health 60(2): 146–148. [DOI] [PMC free article] [PubMed] [Google Scholar]
  111. Woolf SH, Johnson RE, Phillips RL Jr and Philipsen M (2007). "Giving everyone the health of the educated: an examination of whether social change would save more lives than medical advances." American Journal of Public Health 97(4): 679–683. [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

1
NIHMS1653471-supplement-1.docx (692.8KB, docx)

RESOURCES