Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Elevated C-Reactive Protein in Children from Risky Neighborhoods: Evidence for a Stress Pathway Linking Neighborhoods and Inflammation in Children

  • Stephanie T. Broyles ,

    Affiliation Pennington Biomedical Research Center, Louisiana State University System, Baton Rouge, Louisiana, United States of America

  • Amanda E. Staiano,

    Affiliation Pennington Biomedical Research Center, Louisiana State University System, Baton Rouge, Louisiana, United States of America

  • Kathryn T. Drazba,

    Affiliation Pennington Biomedical Research Center, Louisiana State University System, Baton Rouge, Louisiana, United States of America

  • Alok K. Gupta,

    Affiliation Pennington Biomedical Research Center, Louisiana State University System, Baton Rouge, Louisiana, United States of America

  • Melinda Sothern,

    Affiliation School of Public Health, Louisiana State University Health Sciences Center, New Orleans, Louisiana, United States of America

  • Peter T. Katzmarzyk

    Affiliation Pennington Biomedical Research Center, Louisiana State University System, Baton Rouge, Louisiana, United States of America

Elevated C-Reactive Protein in Children from Risky Neighborhoods: Evidence for a Stress Pathway Linking Neighborhoods and Inflammation in Children

  • Stephanie T. Broyles, 
  • Amanda E. Staiano, 
  • Kathryn T. Drazba, 
  • Alok K. Gupta, 
  • Melinda Sothern, 
  • Peter T. Katzmarzyk



Childhood socioeconomic status is linked to adult cardiovascular disease and disease risk. One proposed pathway involves inflammation due to exposure to a stress-inducing neighborhood environment. Whether CRP, a marker of systemic inflammation, is associated with stressful neighborhood conditions among children is unknown.

Methods and Results

The sample included 385 children 5–18 years of age from 255 households and 101 census tracts. Multilevel logistic regression analyses compared children and adolescents with CRP levels >3 mg/L to those with levels ≤3 mg/L across neighborhood environments. Among children living in neighborhoods (census tracts) in the upper tertile of poverty or crime, 18.6% had elevated CRP levels, in contrast to 7.9% of children living in neighborhoods with lower levels of poverty and crime. Children from neighborhoods with the highest levels of either crime or poverty had 2.7 (95% CI: 1.2–6.2) times the odds of having elevated CRP levels when compared to children from other neighborhoods, independent of adiposity, demographic and behavioral differences.


Children living in neighborhoods with high levels of poverty or crime had elevated CRP levels compared to children from other neighborhoods. This result is consistent with a psychosocial pathway favoring early development of cardiovascular risk that involves chronic stress from exposure to socially- and physically-disordered neighborhoods characteristic of poverty.


In adults, low socioeconomic status (SES) has been consistently linked to higher cardiovascular disease (CVD) risk. [1], [2] Similarly, growing evidence links low SES to prevalence of the metabolic syndrome (MetS) [3], [4], [5], [6], [7] and incidence of Type II diabetes (diabetes), [8], [9], [10], [11], [12] risk factors for CVD. Low SES may influence disease risk through behavioral pathways involving poor diet, [13] physical inactivity, [14] and smoking, [15] which often begin in childhood. These behaviors are also associated with low childhood SES. [16], [17] Thus, it is not surprising that low childhood SES has also been implicated in the development of adult metabolic syndrome [7], diabetes [10], and CVD [18], [19].

As an alternative to behavioral pathways, stress has been proposed as a link between poverty and CVD risk. [20], [21], [22] Chronic stress promotes dysregulation of the autonomic nervous system and the hypothalamic-pituitary-adrenal axis, and the resulting inflammation is linked to obesity and the metabolic syndrome. [23], [24], [25], [26] Poverty is often characterized by more frequent exposures to psychosocial stress, including living in socially- and physically-disordered environments. [20], [27], [28], [29], [30] Therefore, exposure to a stressful neighborhood environment may be an important aspect of poverty’s influence on CVD risk. Consistent with this pathway, studies in adults have noted relationships between obesity, [31] diabetes incidence, [32] and heart disease [33], [34] with negative aspects of the neighborhood environment (negative perceptions, [32] neighborhood unemployment and crime, [33], and an index of neighborhood psychosocial stress that includes area SES in its measure [31], [34]).

C-reactive protein (CRP) is a marker of systemic inflammation, and increases in levels of CRP are thought to be part of the cascade of biological responses to chronic stress. [25], [35], [36], [37]. Elevated serum CRP has been found in adults with prediabetes, prehypertension, obesity, diabetes, hypertension and CVD. [38], [39], [40], [41], [42] Childhood CRP levels correlate with CVD risk factors [43], [44] and track into adulthood. [45].

Several studies in adults have noted significant relationships between CRP and neighborhood SES. [46], [47], [48], [49] A growing body of literature has documented links between childhood adversity (measured at the individual or household level) and inflammation; [50] however, whether neighborhood sources of stress are linked to inflammation or other markers of early CVD risk in children has not been well-studied. To address this gap, we examined whether children living in neighborhoods having high levels of poverty or crime (i.e., higher exposure to neighborhood sources of psychosocial stress) had higher levels of CRP than children living in neighborhoods with lower levels of poverty or crime.


Ethics Statement

All study procedures were approved by the Pennington Biomedical Research Center institutional review board, and the parents of the participants provided signed informed consent, with children providing verbal assent.


Participants included children aged 5–18 years recruited for an assessment of factors related to abdominal adiposity. Recruitment occurred through study advertisements (television and print) targeting the Baton Rouge, Louisiana metropolitan area, as well as through pediatricians’ offices. Recruitment attempted to balance the sample across race, sex, and body mass index (BMI) categories over the course of the study. Four-hundred twenty-three (423) children and adolescents participated in the study, which enrolled participants from February 2010 through August 2011.

Of the 423 participants, 9 provided address information that could not be geocoded (e.g., post office boxes). Of the remaining 414 participants, 14 refused to provide a blood sample and, therefore, were missing information on the outcome variable. Participants with CRP>10 mg/L, indicating current or recent acute infection (n = 15), were excluded in order to ensure that the elevated CRP was of non-infectious origin. [51] The final analytic sample consisted of 385 children.

Biological Measures

Fasting serum CRP levels were measured with a high-sensitivity chemiluminescent immunoassay (Siemens Immulite 2000; Deerfield, IL); the lower limit of detection was 0.20 mg/L (37.7% of the sample). The interassay coefficient of variation was 3.7%. Body fat percentage was measured by dual-energy x-ray absorptiometry (DXA) (Hologic QDR 4500A; Bedford, MA).

Neighborhood Measures

Participant addresses were geocoded to the census tract level and linked to census-tract family poverty (US Census 2000) and an index of total crime derived from Uniform Crime Report data (CrimeRisk, Applied Geographic Solutions, 2010). The 2010 CrimeRisk index included data for the years 1998–2006, supplemented with preliminary 2007 release data. The index is adjusted for population and scaled to be relative to the national index of 100. For example, a CrimeRisk index of 150 indicates a crime risk 150% the national average. Within the sample, census tract poverty and crime levels were categorized into tertiles (low, medium, and high).

Other Covariates

A study questionnaire was used to obtain self-reported data on additional covariates. Children completed the survey, with the help of the accompanying parent when necessary. For children under the age of 10, the accompanying parent generally completed the survey.

Household socioeconomic status.

To address the fact that measures of household socioeconomic status were highly correlated with each other and with race, a single household socioeconomic status factor was created using principal components analysis, controlling for race. This factor (Cronbach’s α = 0.70) combined the effects of household poverty income ratio (PIR), father’s educational attainment, and mother’s educational attainment. PIR was created based on the 2009 Federal Poverty Threshold, [52] using self-reported household income (8 categories; $20,000 increments from <$10,000 to >$140,000) and household size. For the household income category that was reported, the median value within the category was used as the value of household income in calculating PIR; for the two extreme categories, the threshold values of $10,000 or $140,000 were used. Father and mother educational attainment was self-reported as grades 0–8, some high school, high school diploma/GED, 1–3 years college, college degree, or post graduate degree.

Self-reported diet and physical activity.

Children reported on levels of physical activity based on a physical activity screening question (days in the past week with at least 60 minutes of physical activity that increased your heart rate and made you breathe hard some of the time) [53] used by both the US Youth Risk Behavior Survey (YRBS) and the Health Behavior in School-Aged Children (HSBC) survey. Children reported on usual intake of various food items, with response options of never, less than once a week, once a week, 2–4 days a week, 5–6 days a week, once a day/every day, or every day more than once. Measures of fruit and vegetable consumption (daily consumption versus less), sweets/sugar-sweetened beverage (SSB) consumption (consumed <2 days per week versus more), and fish consumption (consumed ≥2 days per week versus less) were summed to create an index of a healthy dietary pattern, based on one used to track attainment of the American Heart Association’s criteria for ideal cardiovascular health. [54].

Treatment of Missing Data

Thirty-five participants (9.1%) were missing data on adiposity (n = 4), a household socioeconomic status measure (n = 28), or a behavioral measure (n = 3). Each variable was missing <5%; father’s educational attainment was missing most often, at 4.9%. Participants missing data for any of the covariates were similar to those with complete covariate data with respect to CRP level, race, sex, age, household SES, adiposity, and neighborhood characteristics.

Missing values were multiply-imputed (5 imputations) for these participants using Markov chain Monte Carlo (MCMC) methods, which accommodates non-monotone missing data patterns, under missing at random (MAR) assumptions [55] and using SAS version 9.3 (PROC MI). The imputation model contained all variables included in the full model (Model 3), with the addition of height, which improved imputation of body fat percentage. Although all variables were not normally-distributed, MCMC multiple imputation is robust to departures from normality when the amount of missing data for a particular variable is small. [56] Results across the five imputed datasets were averaged, and the standard errors were adjusted appropriately, using the MIANALYZE procedure in SAS. Sensitivity analyses also examined results from analyses of participants with complete data, and results were similar.


A series of multilevel, multivariable logistic regression analyses (SAS version 9.3, PROC GLIMMIX) was used to examine the relationship between the neighborhood environment and elevated CRP. Because of the large proportion (37.7%) of participants at the lower detection limit, a linear regression model was inappropriate; consequently, CRP was dichotomized as >3 mg/L to those with levels ≤3 mg/L, consistent with other research in children [57], [58] and the definition of “high risk” in adults. [59] Models accounted for both family and neighborhood clustering and compared children and adolescents with CRP levels >3 mg/L to those with levels ≤3 mg/L. Model 1 included the covariates of race (African American versus non-African American), sex, age, and household SES. Because CRP levels are strongly associated with adiposity in children, [60], [61], [62], [63] Model 2 also included a direct measure of adiposity, percent body fat. Model 3 contained the covariates from Model 2, with the addition of physical activity and diet. Each series of models was run according to the following specificationss of neighborhood crime and poverty: I) neighborhood poverty (continuous, centered, standardized) alone in a model, II) neighborhood crime (continuous, centered, standardized) alone in a model, III) neighborhood poverty and crime (continuous, centered, standardized) and their interaction, IV) neighborhood poverty at each tertile of crime, V) neighborhood crime at each tertile of poverty, and VI) neighborhood poverty and crime dichotomized as high neighborhood crime or poverty versus low/medium crime and poverty.

Interaction effects between model covariates were tested where warranted (i.e., when main effects were significant); however, none were identified. Interaction effects were also used to test for sex- and race-based differences in the relationship between elevated CRP levels and the neighborhood environment. These relationships did not differ significantly by race or by sex in any of the models; therefore, overall results are presented.


After excluding observations with CRP levels indicative of a current or recent acute infection, 13.3% of children had CRP levels >3 mg/L. Among children living in neighborhoods (census tracts) with high levels of poverty or crime, 18.6% had elevated CRP levels, in contrast to 7.9% of children living in neighborhoods with lower (low or medium) levels of poverty and crime (Figure 1 and Table 1).

Figure 1. Variation in elevated C-reactive protein concentrations across neighborhood (census tract) poverty and crime.

Study participants represented 255 households from 101 census tracts, 75% of which were located in East Baton Rouge Parish (county), Louisiana. Across all census tracts represented in the sample, the percent of families living in poverty ranged from 0% to 44.2%, with a mean of 13.0% (Table 2). An index of total crime ranged from 6 to 454, with a mean of 202.3. In our sample, the cutoff for high poverty and high crime (the upper tertile within the sample) corresponded to levels of 16.4% and 279, respectively. Poverty and crime were correlated (r = 0.23, p<0.0001). However, when separated into low, medium, and high levels of poverty or crime based on tertiles within the sample, 34 (34%) census tracts had disparate levels of poverty and crime: 17 (17%) census tracts were considered high poverty but low or medium crime, and 17 (17%) were considered high crime but low or medium poverty.

Neither poverty nor crime was associated with elevated CRP when considered singly in a model (Table 3: Models I and II); however, these neighborhood effects interacted significantly (p = 0.02, Models III:1–3). In a model that included main effects for both neighborhood crime and poverty as well as their interaction, both neighborhood crime and poverty were positively associated with elevated CRP; however, as levels of crime or poverty increased, the effect of the other was dampened, as evidenced by an interaction effect <1.0. When poverty was considered within levels of crime (Table 3: Model IV), the effect of poverty on risk for elevated CRP was strongest in low crime areas, and the effect of poverty on CRP decreased significantly with increasing crime (p for trend <0.01 for Models IV:1–3). However, when crime was considered within levels of poverty (Table 3: Model V), the effect of crime on CRP was similar across all levels of poverty. When neighborhood poverty and crime were combined into a single dichotomous term (high neighborhood poverty or crime vs. low or medium poverty and crime), children from neighborhoods with the highest levels of either crime or poverty had 2.7 (95% CI: 1.2–6.2) times the odds of having high CRP levels when compared to children from other neighborhoods, independent of adiposity, demographic and behavioral differences.

Table 3. Odds ratios and 95% confidence intervals for elevated CRP (>3 mg/L) associated with neighborhood crime and poverty.

Across all analytic specifications of neighborhood crime or poverty, inclusion of adiposity in the model (Model 2 vs. Model 1) generally attenuated the association between neighborhood crime/poverty and CRP. Further inclusion of behavioral variables (Model 3 vs. Model 2) did not appear to modify the estimated relationships.


In a cross-sectional sample of children ages 5 to 18 years, children from neighborhoods characterized by high levels of poverty or crime exhibited higher levels of CRP, a marker of systemic inflammation and cardiovascular risk that tracks into adulthood. [45] The life course model for disease development suggests that risk due to exposure to adverse environments accumulates over a person's lifetime, beginning in childhood. [18] Although considerable evidence links childhood SES with adult CVD, the mechanisms for this association remain poorly understood. Our findings suggest that inflammation due to exposure to a stress-inducing neighborhood environment may be one pathway.

A recent review noted that stress may be a factor on par with diet and physical activity in the development of obesity and related metabolic disease. [64] Our research suggests that children living in neighborhoods with high levels of poverty or crime may be at increased risk of inflammation resulting from exposure to stressful neighborhood conditions. This result is consistent with others in adults that noted significant relationships between CRP and neighborhood SES, [46], [47], [48], [49] between abnormal glucose metabolism and area SES, [65] and between heart disease and neighborhood unemployment and crime [33] or an index of neighborhood psychosocial stress. [34] More numerous are studies noting an association between individual-level measures of adversity, like household SES, and inflammation in children (reviewed in [50]) or between household SES and CRP in adults [49], [66], [67], [68], [69] or other biomarkers of CVD risk in adolescents. [70] However, results are mixed, and some associations are attenuated after adjustment for adiposity. [69], [71], [72], [73] Furthermore, recent work has noted an association between neighborhood disorder and serum cortisol levels in children, [74] providing additional support for a neighborhood stress-inflammation relationship in children.

Results further suggest that neighborhood crime and poverty are not independent in their association with elevated CRP. Specifically, the effect of poverty was strongest at lower levels of crime and decreased with increasing crime, and the effect of crime was strongest at lower levels of poverty and decreased with increasing poverty. While neither crime nor poverty appeared to be associated with elevated CRP when considered individually in a model, associations were apparent when the analysis focused on participants living in the low and medium tertiles of the other neighborhood effect. Children living in the highest tertile of crime or poverty experienced the highest risk of elevated CRP; however, within these groups, exposure to an additional neighborhood stressor did not appear to alter risk (Figure 2). Consequently, children exposed to either high neighborhood crime or high neighborhood poverty appear to be at highest risk of elevated CRP. Studies of the influence of the neighborhood environment have cited the importance of “context” in understanding health and health behavior. [75] Our results further suggest that even when studying a neighborhood effect, the expanded context of the neighborhood may matter. Studies of neighborhood effects typically investigate effects singly, and our results suggest that important effects may be missed, or their effect sizes attenuated, by such an approach.

Figure 2. Covariate-adjusted elevated C-reactive protein (CRP) concentrations across neighborhood (census tract) poverty and crime levels.

(A) Percent of children with elevated CRP in low, medium, and high crime neighborhoods, by low-medium poverty versus high poverty neighborhoods, and (B) percent of children with elevated CRP in low, medium, and high poverty neighborhoods, by low-medium crime versus high crime neighborhoods.

In the current study, the effect size seen for high poverty or crime neighborhood and elevated CRP is estimated to be 0.55, [76] which is considered a medium effect. [77] Neighborhood effects on children have typically been small; [78] although, there are examples of studies with similarly-sized effects. In a study looking at the relative contribution of neighborhood and household measures of SES, neighborhood income and education levels were stronger than their household equivalents in adjusted analyses of BMI and cortisol level. [79] Neighborhood SES independently accounted for 9.2–10.6% of the variance in BMI, equivalent to a moderate effect size. In another study that looked at area effects on fibrinogen, [80] the effect size noted, while apparently small for a unit change in the deprivation index used (0.05 SD per unit change), could approach a medium effect size if children from the upper and lower proportions of the neighborhood deprivation range (an absolute range of 15.1 units in the sample) had been compared. Lastly, a study of neighborhood SES and cardiovascular responsivity to and recovery from laboratory stressors found significant associations in both white and African American children. [81] Both heart rate responsivity and recovery showed medium effects between the upper and lower neighborhood SES groups in both white (−0.58 SD responsivity effect and −0.46 SD recovery effect) and African American children (0.52 SD responsivity effect and 0.62 SD recovery effect), although the direction of the effect differed between the two race groups.

While not the focus of this study, household SES was not related to elevated CRP in our analyses. Although research in adults has documented inverse relationships between markers of socioeconomic status and CRP levels, [67], [68] this relationship has been inconsistent in children. [50] Even among studies reporting associations, the associations generally disappear after adjustment for adiposity. [69], [72], [73] It is not entirely clear why inflammation in children would be more strongly related to neighborhood conditions, versus household SES, and we look to future studies to confirm our results. However, a study investigating the joint contributions of family and neighborhood SES on health markers in adolescents found that neighborhood SES was more strongly related to BMI and basal cortisol levels than family SES. [79] Other research reporting associations between area conditions and cortisol [74] or fibrinogen [80] did not adjust for household SES.

Studies of CRP in children have consistently documented a strong relationship between CRP and indirect measures of adiposity, [60], [61], [62], [63] with BMI or derivative measures (e.g., BMI percentiles or BMI categories) being the most common. In the current study, a direct measure of adiposity – body fat percentage, assessed by DXA – was used. DXA is considered to provide the most accurate body composition analysis in children. [82] While BMI is the most common metric for assessing adiposity, it in fact measures excess body mass, which can be composed of either lean or fat mass, and its accuracy in measuring adiposity varies according to a child’s actual adiposity (assessed by DXA). [83], [84] While not reported, among all covariates, body fat percentage showed the strongest association with elevated CRP across all models. Additionally, all results were consistent when sex- and age-specific BMI z-scores were substituted for body fat percentage in the analysis.

CRP levels become elevated after infection, and inverse relationships between SES and illness have been noted. [85] Moreover, elevated CRP levels are associated with obesity in children, [60], [61], [62], [63] which has also been linked to both individual/family and neighborhood SES. [86], [87], [88], [89] A particular strength of this study is its ability to focus on the effect of interest, i.e., the proposed pathway whereby the neighborhood environment promotes chronic stress and elevated CRP levels, as analyses were adjusted for a direct measure of adiposity and other confounding relationships. Disentangling neighborhood effects of SES on health is methodologically challenging. Our use of multilevel models that account for within-household similarities and shared environment, including psychosocial stressors in the household, strengthen the evidence for the neighborhood effect.

The cross-sectional study design limits our findings, as these results cannot demonstrate causative relationships. Furthermore, participants self-selected into the study, and recruitment attempted to balance across race, sex, and BMI categories; therefore, the participant sample cannot be considered demographically representative of the geographic area in which study recruitment occurred. Study participants did represent 76 (83%) of the census tracts within the parish (county) in which the majority of participants resided, however. Also, census tracts in which study participants resided were generally similar to those not represented in the sample.

Our results may also be limited by the presence of missing data. A sensitivity analysis did not reveal differences between analyses of complete cases and results reported here; however, neither complete case analysis nor multiple imputation can correct for the 23 participants excluded from the analysis because of missing exposure (address not able to be geocoded) or missing outcome (refusal of blood draw). Our results could be biased if these excluded participants differed in the adjusted relationship between their neighborhood conditions and CRP levels, although the effect is likely to be minimal given the small number of exclusions.

A large proportion of study participants (37.7%) had CRP levels at the lower limit of detection, precluding analysis of CRP as a continuous outcome, as has been done in other studies. Additionally, there are no specific pediatric guidelines on a risk cut-off for CRP levels in children. [44] Other studies in CRP in children, however, have applied the adult criteria to define an dichotomous outcome of elevated CRP. [58], [90] Although our choice of cut-off reflects a preference for a criterion-based cutoff, we did examine others to ensure that results were not sensitive to the choice of cut-off. Generally, the effect size (odds ratio) was smaller with a more inclusive definition of elevated CRP (e.g., when defined as the top quartile, CRP levels ≥1.2 mg/L were considered “elevated”); however, the association with neighborhood crime/poverty remained significant.

Results may also be related to unmeasured similarities in families that live in the high poverty/crime neighborhoods. As an example, if children in the different neighborhood conditions were differentially exposed to second-hand smoke, this would not have been controlled for in our analyses. However, studies have not found relationships with second-hand smoke exposure and CRP, after adjustment for adiposity, [60], [69] so this particular example is unlikely to have confounded our results. Furthermore, there may be individual level factors not accounted for in the analysis, like smoking and alcohol consumption. Although we did assess both self-reported smoking and alcohol consumption, only 4 children (1.0%) reported smoking and only 5 children (1.3%) reported alcohol consumption. While neither behavior was associated with elevated CRP in our sample, we recognize that these behaviors may have been underreported and may represent unmeasured confounders of the relationship being investigated.

Lastly, although we confirmed a significant relationship between CRP and neighborhood poverty/crime, we did not include any direct psychological measures of stress; therefore, these results cannot confirm the existence of a neighborhood environment-psychological stress-CRP pathway.

In conclusion, children living in neighborhoods with high levels of poverty or crime had elevated CRP levels compared to children from neighborhoods with lower poverty and crime. This result is consistent with a psychosocial pathway favoring the early development of cardiovascular risk that involves chronic stress from exposure to social and physical disorder characteristic of impoverished neighborhoods. Cardiovascular disease pathways involving neighborhood stress may initiate in childhood. Thus, prevention and early disease screening may have maximal impact when targeting children living in neighborhoods with high levels of poverty or crime.

Author Contributions

Conceived and designed the experiments: STB PTK AKG. Performed the experiments: STB PTK AKG. Analyzed the data: STB. Wrote the paper: STB. Collection of data: AES KTD. Interpretation of data: AES KTD AKG MS PTK. Critical review of manuscript: AES KTD AKG MS PTK. Final approval of manuscript: STB AES KTD AKG MS PTK.


  1. 1. Kaplan GA, Keil JE (1993) Socioeconomic factors and cardiovascular disease: a review of the literature. Circulation 88: 1973–1998. doi: 10.1161/01.cir.88.4.1973
  2. 2. Clark AM, DesMeules M, Luo W, Duncan AS, Wielgosz A (2009) Socioeconomic status and cardiovascular disease: risks and implications for care. Nat Rev Cardiol 6: 712–722. doi: 10.1038/nrcardio.2009.163
  3. 3. Brunner EJ, Marmot MG, Nanchahal K, Shipley MJ, Stansfeld SA, et al. (1997) Social inequality in coronary risk: central obesity and the metabolic syndrome. Evidence from the Whitehall II study. Diabetologia 40: 1341–1349. doi: 10.1007/s001250050830
  4. 4. Dallongeville J, Cottel D, Ferrieres J, Arveiler D, Bingham A, et al. (2005) Household income is associated with the risk of metabolic syndrome in a sex-specific manner. Diabetes Care 28: 409–415. doi: 10.2337/diacare.28.2.409
  5. 5. Perel P, Langenberg C, Ferrie J, Moser K, Brunner E, et al. (2006) Household wealth and the metabolic syndrome in the Whitehall II study. Diabetes Care 29: 2694–2700. doi: 10.2337/dc06-0022
  6. 6. Chichlowska KL, Rose KM, Diez-Roux AV, Golden SH, McNeill AM, et al. (2008) Individual and neighborhood socioeconomic status characteristics and prevalence of metabolic syndrome: the Atherosclerosis Risk in Communities (ARIC) Study. Psychosom Med 70: 986–992. doi: 10.1097/psy.0b013e318183a491
  7. 7. Chichlowska KL, Rose KM, Diez-Roux AV, Golden SH, McNeill AM, et al. (2009) Life course socioeconomic conditions and metabolic syndrome in adults: the Atherosclerosis Risk in Communities (ARIC) Study. Ann Epidemiol 19: 875–883. doi: 10.1016/j.annepidem.2009.07.094
  8. 8. Kumari M, Head J, Marmot M (2004) Prospective study of social and other risk factors for incidence of type 2 diabetes in the Whitehall II study. Arch Intern Med 164: 1873–1880. doi: 10.1001/archinte.164.17.1873
  9. 9. Robbins JM, Vaccarino V, Zhang H, Kasl SV (2005) Socioeconomic status and diagnosed diabetes incidence. Diabetes Res Clin Pract 68: 230–236. doi: 10.1016/j.diabres.2004.09.007
  10. 10. Lidfeldt J, Li TY, Hu FB, Manson JE, Kawachi I (2007) A prospective study of childhood and adult socioeconomic status and incidence of type 2 diabetes in women. Am J Epidemiol 165: 882–889. doi: 10.1093/aje/kwk078
  11. 11. Williams ED, Tapp RJ, Magliano DJ, Shaw JE, Zimmet PZ, et al. (2010) Health behaviours, socioeconomic status and diabetes incidence: the Australian Diabetes Obesity and Lifestyle Study (AusDiab). Diabetologia 53: 2538–2545. doi: 10.1007/s00125-010-1888-4
  12. 12. Lee TC, Glynn RJ, Pena JM, Paynter NP, Conen D, et al. (2011) Socioeconomic status and incident type 2 diabetes mellitus: data from the Women's Health Study. PLoS One 6: e27670. doi: 10.1371/journal.pone.0027670
  13. 13. Darmon N, Drewnowski A (2008) Does social class predict diet quality? Am J Clin Nutr 87: 1107–1117.
  14. 14. Trost SG, Owen N, Bauman AE, Sallis JF, Brown W (2002) Correlates of adults' participation in physical activity: review and update. Med Sci Sports Exerc 34: 1996–2001. doi: 10.1097/00005768-200212000-00020
  15. 15. Garrett BE, Dube SR, Trosclair A, Caraballo RS, Pechacek TF (2011) Cigarette smoking - United States, 1965–2008. MMWR Surveill Summ 60 Suppl: 109–113
  16. 16. Lee RE, Cubbin C (2002) Neighborhood context and youth cardiovascular health behaviors. American Journal of Public Health 92: 428–436. doi: 10.2105/ajph.92.3.428
  17. 17. Hanson M, Chen E (2007) Socioeconomic status and health behaviors in adolescence: a review of the literature. Journal of Behavioral Medicine 30: 263–285. doi: 10.1007/s10865-007-9098-3
  18. 18. Kuh D, Ben-Shlomo Y, editors (2004) A life course approach to chronic disease epidemiology. New York: Oxford University Press.
  19. 19. Galobardes B, Smith GD, Lynch JW (2006) Systematic review of the influence of childhood socioeconomic circumstances on risk for cardiovascular disease in adulthood. Annals of Epidemiology 16: 91–104. doi: 10.1016/j.annepidem.2005.06.053
  20. 20. Brunner E (1997) Stress and the biology of inequality. BMJ 314: 1472–1476. doi: 10.1136/bmj.314.7092.1472
  21. 21. Baum A, Garofalo JP, Yali AM (1999) Socioeconomic status and chronic stress. Does stress account for SES effects on health? Ann N Y Acad Sci 896: 131–144. doi: 10.1111/j.1749-6632.1999.tb08111.x
  22. 22. McEwen BS, Seeman T (1999) Protective and damaging effects of mediators of stress. Elaborating and testing the concepts of allostasis and allostatic load. Ann N Y Acad Sci 896: 30–47. doi: 10.1111/j.1749-6632.1999.tb08103.x
  23. 23. Bjorntorp P (2001) Do stress reactions cause abdominal obesity and comorbidities? Obes Rev 2: 73–86. doi: 10.1046/j.1467-789x.2001.00027.x
  24. 24. Kassi E, Pervanidou P, Kaltsas G, Chrousos G (2011) Metabolic syndrome: definitions and controversies. BMC Med 9: 48. doi: 10.1186/1741-7015-9-48
  25. 25. Black PH, Garbutt LD (2002) Stress, inflammation and cardiovascular disease. J Psychosom Res 52: 1–23. doi: 10.1016/s0022-3999(01)00302-6
  26. 26. Brunner EJ, Hemingway H, Walker BR, Page M, Clarke P, et al. (2002) Adrenocortical, autonomic, and inflammatory causes of the metabolic syndrome: nested case-control study. Circulation 106: 2659–2665. doi: 10.1161/
  27. 27. Attar BK, Guerra NG, Tolan PH (1994) Neighborhood disadvantage, stressful life events, and adjustment in urban elementary-school-children. Journal of Clinical Child Psychology 23: 391–400. doi: 10.1207/s15374424jccp2304_5
  28. 28. Ross CE, Mirowsky J (2001) Neighborhood Disadvantage, Disorder, and Health. Journal of Health and Social Behavior 42: 258–276. doi: 10.2307/3090214
  29. 29. Steptoe A, Feldman PJ (2001) Neighborhood problems as sources of chronic stress: development of a measure of neighborhood problems, and associations with socioeconomic status and health. Ann Behav Med 23: 177–185. doi: 10.1207/s15324796abm2303_5
  30. 30. Hill TD, Ross CE, Angel RJ (2005) Neighborhood disorder, psychophysiological distress, and health. J Health Soc Behav 46: 170–186. doi: 10.1177/002214650504600204
  31. 31. Glass TA, Rasmussen MD, Schwartz BS (2006) Neighborhoods and obesity in older adults: the Baltimore Memory Study. American Journal of Preventive Medicine 31: 455–463. doi: 10.1016/j.amepre.2006.07.028
  32. 32. Goldoftas BF (2008) Neighborhood Environmental Health and Risk Factors for Type 2 Diabetes in the English Longitudinal Study of Ageing. Epidemiology 19: S268 210.1097/1001.ede.0000340283.0000300157.0000340287f.
  33. 33. Sundquist K, Theobald H, Yang M, Li X, Johansson SE, et al. (2006) Neighborhood violent crime and unemployment increase the risk of coronary heart disease: a multilevel study in an urban setting. Soc Sci Med 62: 2061–2071. doi: 10.1016/j.socscimed.2005.08.051
  34. 34. Augustin T, Glass TA, James BD, Schwartz BS (2008) Neighborhood psychosocial hazards and cardiovascular disease: the Baltimore Memory Study. Am J Public Health 98: 1664–1670. doi: 10.2105/ajph.2007.125138
  35. 35. McDade TW, Hawkley LC, 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: 376–381. doi: 10.1097/01.psy.0000221371.43607.64
  36. 36. Steptoe A, Hamer M, Chida Y (2007) The effects of acute psychological stress on circulating inflammatory factors in humans: a review and meta-analysis. Brain Behav Immun 21: 901–912. doi: 10.1016/j.bbi.2007.03.011
  37. 37. Gouin JP, Glaser R, Malarkey WB, Beversdorf D, Kiecolt-Glaser J (2011) Chronic stress, daily stressors, and circulating inflammatory markers. Health Psychol.2011/09/21 ed.
  38. 38. Gupta AK, Johnson WD (2010) Prediabetes and prehypertension in disease free obese adults correlate with an exacerbated systemic proinflammatory milieu. J Inflamm 7: 36. doi: 10.1186/1476-9255-7-36
  39. 39. Visser M, Bouter LM, McQuillan GM, Wener MH, Harris TB (1999) Elevated C-reactive protein levels in overweight and obese adults. JAMA 282: 2131–2135. doi: 10.1001/jama.282.22.2131
  40. 40. Pradhan AD, Manson JE, Rifai N, Buring JE, Ridker PM (2001) C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA 286: 327–334. doi: 10.1001/jama.286.3.327
  41. 41. Sesso HD, Buring JE, Rifai N, Blake GJ, Gaziano JM, et al. (2003) C-reactive protein and the risk of developing hypertension. JAMA 290: 2945–2951. doi: 10.1016/j.accreview.2004.02.025
  42. 42. Kaptoge S, Di Angelantonio E, Lowe G, Pepys MB, Thompson SG, et al. (2010) C-reactive protein concentration and risk of coronary heart disease, stroke, and mortality: an individual participant meta-analysis. Lancet 375: 132–140. doi: 10.1016/s0140-6736(09)61717-7
  43. 43. Jarvisalo MJ, Harmoinen A, Hakanen M, Paakkunainen U, Viikari J, et al. (2002) Elevated serum C-reactive protein levels and early arterial changes in healthy children. Arterioscler Thromb Vasc Biol 22: 1323–1328. doi: 10.1161/01.atv.0000024222.06463.21
  44. 44. Balagopal PB, de Ferranti SD, Cook S, Daniels SR, Gidding SS, et al. (2011) Nontraditional risk factors and biomarkers for cardiovascular disease: mechanistic, research, and clinical considerations for youth: a scientific statement from the American Heart Association. Circulation 123: 2749–2769. doi: 10.1161/cir.0b013e31821c7c64
  45. 45. Juonala M, Viikari JS, Ronnemaa T, Taittonen L, Marniemi J, et al. (2006) Childhood C-reactive protein in predicting CRP and carotid intima-media thickness in adulthood: the Cardiovascular Risk in Young Finns Study. Arterioscler Thromb Vasc Biol 26: 1883–1888. doi: 10.1161/01.atv.0000228818.11968.7a
  46. 46. Pollitt RA, Kaufman JS, Rose KM, Diez-Roux AV, Zeng D, et al. (2007) Early-life and adult socioeconomic status and inflammatory risk markers in adulthood. Eur J Epidemiol 22: 55–66. doi: 10.1007/s10654-006-9082-1
  47. 47. Petersen KL, Marsland AL, Flory J, Votruba-Drzal E, Muldoon MF, et al. (2008) Community socioeconomic status is associated with circulating interleukin-6 and C-reactive protein. Psychosom Med 70: 646–652. doi: 10.1097/psy.0b013e31817b8ee4
  48. 48. Pollitt RA, Kaufman JS, Rose KM, Diez-Roux AV, Zeng D, et al. (2008) Cumulative life course and adult socioeconomic status and markers of inflammation in adulthood. J Epidemiol Community Health 62: 484–491. doi: 10.1136/jech.2006.054106
  49. 49. Schafer MH, Ferraro KF, Williams SR (2011) Low socioeconomic status and body mass index as risk factors for inflammation in older adults: conjoint influence on C-reactive protein? Journals of Gerontology Series A-Biological Sciences and Medical Sciences 66: 667–673. doi: 10.1093/gerona/glr009
  50. 50. Slopen N, Koenen KC, Kubzansky LD (2011) Childhood adversity and immune and inflammatory biomarkers associated with cardiovascular risk in youth: A systematic review. Brain Behav Immun.
  51. 51. Jaye DL, Waites KB (1997) Clinical applications of C-reactive protein in pediatrics. Pediatr Infect Dis J 16: 735–746; quiz 746–737.
  52. 52. U.S. Census Bureau (2010) Poverty Thresholds 2009.
  53. 53. Prochaska JJ, Sallis JF, Long B (2001) A physical activity screening measure for use with adolescents in primary care. Arch Pediatr Adolesc Med 155: 554–559. doi: 10.1001/archpedi.155.5.554
  54. 54. Lloyd-Jones DM, Hong Y, Labarthe D, Mozaffarian D, Appel LJ, et al. (2010) Defining and setting national goals for cardiovascular health promotion and disease reduction: the American Heart Association's strategic Impact Goal through 2020 and beyond. Circulation 121: 586–613. doi: 10.1161/circulationaha.109.192703
  55. 55. Little RJA, Rubin DB (1987) Statistical analysis with missing data. New York: J. Wiley & Sons.
  56. 56. Schafer JL (1997) Analysis of Incomplete Multivariate Data. New York: Chapman and Hall.
  57. 57. Ford ES, Ajani UA, Mokdad AH (2005) The metabolic syndrome and concentrations of C-reactive protein among U.S. youth. Diabetes Care 28: 878–881. doi: 10.2337/diacare.28.4.878
  58. 58. Lande MB, Pearson TA, Vermilion RP, Auinger P, Fernandez ID (2008) Elevated blood pressure, race/ethnicity, and C-reactive protein levels in children and adolescents. Pediatrics 122: 1252–1257. doi: 10.1542/peds.2007-3162
  59. 59. Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon RO, 3rd, et al (2003) Markers of inflammation and cardiovascular disease: application to clinical and public health practice: A statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation 107: 499–511. doi: 10.1161/01.cir.0000052939.59093.45
  60. 60. Cook DG, Mendall MA, Whincup PH, Carey IM, Ballam L, et al. (2000) C-reactive protein concentration in children: relationship to adiposity and other cardiovascular risk factors. Atherosclerosis 149: 139–150. doi: 10.1016/s0021-9150(99)00312-3
  61. 61. Ford ES, Galuska DA, Gillespie C, Will JC, Giles WH, et al. (2001) C-reactive protein and body mass index in children: findings from the Third National Health and Nutrition Examination Survey, 1988–1994. J Pediatr 138: 486–492. doi: 10.1067/mpd.2001.112898
  62. 62. Ford ES (2003) C-reactive protein concentration and cardiovascular disease risk factors in children: findings from the National Health and Nutrition Examination Survey 1999–2000. Circulation 108: 1053–1058. doi: 10.1161/01.cir.0000080913.81393.b8
  63. 63. Skinner AC, Steiner MJ, Henderson FW, Perrin EM (2010) Multiple markers of inflammation and weight status: cross-sectional analyses throughout childhood. Pediatrics 125: e801–809. doi: 10.1542/peds.2009-2182
  64. 64. Holmes ME, Ekkekakis P, Eisenmann JC (2010) The physical activity, stress and metabolic syndrome triangle: a guide to unfamiliar territory for the obesity researcher. Obes Rev 11: 492–507. doi: 10.1111/j.1467-789x.2009.00680.x
  65. 65. Williams ED, Magliano DJ, Zimmet PZ, Kavanagh AM, Stevenson CE, et al. (2012) Area-Level Socioeconomic Status and Incidence of Abnormal Glucose Metabolism: The Australian Diabetes, Obesity and Lifestyle (AusDiab) study. Diabetes Care 35: 1455–1461. doi: 10.2337/dc11-1410
  66. 66. Owen N, Poulton T, Hay FC, Mohamed-Ali V, Steptoe A (2003) Socioeconomic status, C-reactive protein, immune factors, and responses to acute mental stress. Brain, Behavior, and Immunity 17: 286–295. doi: 10.1016/s0889-1591(03)00058-8
  67. 67. Nazmi A, 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: 10.1186/1471-2458-7-212
  68. 68. Muennig P, Sohler N, Mahato B (2007) Socioeconomic status as an independent predictor of physiological biomarkers of cardiovascular disease: Evidence from NHANES. Preventive Medicine 45: 35–40. doi: 10.1016/j.ypmed.2007.04.005
  69. 69. Dowd JB, Zajacova A, Aiello AE (2010) Predictors of inflammation in U.S. children aged 3–16 years. Am J Prev Med 39: 314–320. doi: 10.1016/j.amepre.2010.05.014
  70. 70. Goodman E, McEwen BS, Huang B, Dolan LM, Adler NE (2005) Social inequalities in biomarkers of cardiovascular risk in adolescence. Psychosom Med 67: 9–15. doi: 10.1097/01.psy.0000149254.36133.1a
  71. 71. Alley DE, Seeman TE, Ki Kim J, Karlamangla A, Hu P, et al. (2006) Socioeconomic status and C-reactive protein levels in the US population: NHANES IV. Brain Behav Immun 20: 498–504. doi: 10.1016/j.bbi.2005.10.003
  72. 72. Howe LD, Galobardes B, Sattar N, Hingorani AD, Deanfield J, et al. (2010) Are there socioeconomic inequalities in cardiovascular risk factors in childhood, and are they mediated by adiposity? Findings from a prospective cohort study. International Journal of Obesity 34: 1149–1159. doi: 10.1038/ijo.2010.52
  73. 73. Murasko JE (2008) Male-female differences in the association between socioeconomic status and atherosclerotic risk in adolescents. Soc Sci Med 67: 1889–1897. doi: 10.1016/j.socscimed.2008.09.018
  74. 74. Dulin-Keita A, Casazza K, Fernandez JR, Goran MI, Gower B (2012) Do neighbourhoods matter? Neighbourhood disorder and long-term trends in serum cortisol levels. J Epidemiol Community Health 66: 24–29. doi: 10.1136/jech.2009.092676
  75. 75. Link BG, Phelan J (1995) Social conditions as fundamental causes of disease. Journal of Health and Social Behavior Extra Issue: 80–94.
  76. 76. Chinn S (2000) A simple method for converting an odds ratio to effect size for use in meta-analysis. Stat Med 19: 3127–3131. doi: 10.1002/1097-0258(20001130)19:22<3127::aid-sim784>;2-m
  77. 77. Cohen J (1992) A power primer. Psychol Bull 112: 155–159. doi: 10.1037//0033-2909.112.1.155
  78. 78. Leventhal T, Brooks-Gunn J (2000) The neighborhoods they live in: The effects of neighborhood residence on child and adolescent outcomes. Psychological Bulletin 126: 309–337. doi: 10.1037/0033-2909.126.2.309
  79. 79. Chen E, Paterson LQ (2006) Neighborhood, family, and subjective socioeconomic status: How do they relate to adolescent health? Health Psychol 25: 704–714. doi: 10.1037/0278-6133.25.6.704
  80. 80. Morley R, Harland P, Law CM, Lucas A (2000) Birthweight and social deprivation: influences on serum lipids and fibrinogen. Acta Paediatr 89: 703–707. doi: 10.1111/j.1651-2227.2000.tb00369.x
  81. 81. Jackson RW, Treiber FA, Turner JR, Davis H, Strong WB (1999) Effects of race, sex, and socioeconomic status upon cardiovascular stress responsivity and recovery in youth. Int J Psychophysiol 31: 111–119. doi: 10.1016/s0167-8760(98)00044-0
  82. 82. Helba M, Binkovitz LA (2009) Pediatric body composition analysis with dual-energy X-ray absorptiometry. Pediatr Radiol 39: 647–656. doi: 10.1007/s00247-009-1247-0
  83. 83. Freedman DS, Wang J, Maynard LM, Thornton JC, Mei Z, et al. (2005) Relation of BMI to fat and fat-free mass among children and adolescents. Int J Obes (Lond) 29: 1–8. doi: 10.1038/sj.ijo.0802735
  84. 84. Federico B, D'Aliesio F, Pane F, Capelli G, Rodio A (2011) Body mass index has a curvilinear relationship with the percentage of body fat among children. BMC Res Notes 4: 301. doi: 10.1186/1756-0500-4-301
  85. 85. Cohen S (1999) Social status and susceptibility to respiratory infections. Ann N Y Acad Sci 896: 246–253. doi: 10.1111/j.1749-6632.1999.tb08119.x
  86. 86. Sundquist J, Malmstrom M, Johansson S-E (1999) Cardiovascular risk factors and the neighbourhood environment: a multilevel analysis. International Journal of Epidemiology 28: 841–845. doi: 10.1093/ije/28.5.841
  87. 87. McLaren L (2007) Socioeconomic status and obesity. Epidemiol Rev 29: 29–48. doi: 10.1093/epirev/mxm001
  88. 88. Ludwig J, Sanbonmatsu L, Gennetian L, Adam E, Duncan GJ, et al. (2011) Neighborhoods, obesity, and diabetes–a randomized social experiment. N Engl J Med 365: 1509–1519. doi: 10.1056/nejmsa1103216
  89. 89. Shrewsbury V, Wardle J (2008) Socioeconomic status and adiposity in childhood: a systematic review of cross-sectional studies 1990–2005. Obesity (Silver Spring) 16: 275–284. doi: 10.1038/oby.2007.35
  90. 90. Fuligni AJ, Telzer EH, Bower J, Cole SW, Kiang L, et al. (2009) A preliminary study of daily interpersonal stress and C-reactive protein levels among adolescents from Latin American and European backgrounds. Psychosom Med 71: 329–333. doi: 10.1097/psy.0b013e3181921b1f