Social Disparities in Biological Risk Factors for Cancer ......Nov 01, 2016  · Datasets across the Life Span Adolescence and Young adulthood: National Longitudinal Study of Adolescent

Yang Claire Yang

P r o f e s s o r , D e p a r t m e n t o f S o c i o l o g yL i n e b e r g e r C o m p r e h e n s i v e C a n c e r C e n t e r ( L C C C )C a r o l i n a P o p u l a t i o n C e n t e r ( C P C )

Presentation at the Cancer Outcomes Breakfast Seminar, November 1st, 2016

Social Disparities in Biological Risk Factors for Cancer in Young Adulthood: Obesity, Inflammation, and Socio-behavioral Mechanisms

Overview

Temporal trends and population heterogeneity in cancer incidence & mortality

Early-life social status and adulthood inflammation and metabolic disorder

Social relationships, inflammation, and cancer survival

Social disparities in obesity and inflammation in young adults: behavioral mechanisms

Temporal Dynamics and Cohort Analysis

Conceptualization: Cohort Effects Cohort effects are changes across groups of individuals who

experience an initial event such as birth in the same year or years, reflecting: Early life conditions Life-long cumulative exposures to social, behavioral, and

biological risk factors Life course context of individual aging process.

Cohort Analysis Goal: distinguish age, period, and cohort variations New methods: the generalized linear mixed models (Yang and

Land 2013) http://yangclaireyang.web.unc.edu/age-period-cohort-analysis-new-models-methods-and-empirical-applications

http://yangclaireyang.web.unc.edu/age-period-cohort-analysis-new-models-methods-and-empirical-applications/

Figure 1. Estimated Cohort Effects of Cancer Mortality by Sex and Race

Source: Yang and Land (2013). (Data source: NCHS/CDC vital statistics: 1969 – 2002)

Figure 2. Major Behavioral Risk Factors: Cohort Patterns of Smoking

Source: Yang and Land (2013). (Data source: National Health Interview Survey)

Figure 3. Estimated Cohort Effects of Colorectal Cancer Incidence

Source: Yang and Land (2013). (Data source: SEER 1973 - 2008)

Figure 4. Predicted Incidence of Colon Cancer by Age, 2010-2030

Source: Murphy, C. 2016. Adapted from: Bailey CE et al., JAMA Surg 2015;150(1):17-22

Figure 5. Major Behavioral Risk Factors: Cohort Effects of Obesity

Source: Yang and Land (2013). (Data source: National Health and Nutrition Examination Survey: 1971 – 2007)

Figure 6. Major Behavioral Risk Factors: Cohort Patterns of Diet

Source: Reither, Yang, Robinson, and Ng (work in progress). (Data source: National Health and Nutrition Examination Survey: 1971 – 2007)

A General Life Course Model of Social Disparities in Health

Source: Yang et al. (2016): PNAS

Social Stress Chronic Disease

InflammationMetabolic Syndrome

Allostatic Load

A

C B

D

Childhood Adolescence Young Adulthood Mid Adulthood Late Adulthood

1. Early-life Social Status and Adulthood Inflammation and Metabolic Disorder

1 . 1 Y a n g e t a l . ( 2 0 1 5 ) : R e v i s e a n d r e s u b m i t1 . 2 Y a n g e t a l . ( 2 0 1 6 ) : U n d e r r e v i e w

(Nature 2002) (Hussain and Harris 2007: Int. J. Cancer)

(Marx 2004: Science)

Datasets across the Life Span

Adolescence and Young adulthood: National Longitudinal Study of Adolescent to Adult Health (Add

Health) Wave I (1994-95) Age 12-18; Wave IV (2008-09) Age 24-32 N = 12,237

Young/mid adulthood: National Survey of Midlife Development in the United States

(MIDUS) Wave I (1995-96) Age 25-64; Wave II (2004-06) Age 34-74 N = 908

Late adulthood: National Social Life, Health, and Aging Project (NSHAP)

Wave I (2005-06) Age 57-85; Wave II (2010-11) Age 63-91 N = 1571

Key Measures: Harmonized across Datasets

Socioeconomic Status (SES) Early-life SES

Parent education, household income, welfare use, subjective financial well-being

Adult SES Education, household income, welfare use, household assets

Biomarkers Inflammation: C-reactive protein (CRP) Metabolic syndrome: blood pressure, waist circumference,

cholesterol

Figure 7. Early-life SES Associated with Biomarkers in Young Adulthood (Add Health): Accumulation of Risks

Note: All models adjust for age, sex, race/ethnicity, marital status, and childhood health.

0

0.2

0.4

0.6

0.8

1

1.2

Inflammation Hypertension Abdominalobesity

Overall obesity

Od

ds

Rat

ios

(95%

CI)

Early life SESonly

Early life SES(adj forcurrent)

Current SES(adj for earlylife)

p

Figure 8. SES Mobility Associated with Biomarkers across the Life Span

Note: +p

2. Social Relationships, Inflammation, and Cancer Survival

2 . 1 Y a n g e t a l . ( 2 0 1 3 ) . J . o f H e a l t h a n d S o c i a l B e h a v i o r2 . 2 Y a n g e t a l . ( 2 0 1 4 ) . B i o d e m o g r a p h y & S o c i a l B i o l o g y2 . 3 Y a n g e t a l . ( 2 0 1 6 ) . P N A S2 . 4 B o e n & Y a n g e t a l . ( I n p r o g r e s s )

Social Isolation and Tumor Incidence: Model Animals

Source: McClintock et al. (2005), Journals of Gerontology: Series B.

Data and Measures

The National Health and Nutrition Examination Survey (NHANES) Linked Mortality File 1988 – 2006; age = 40+; N = 6,729 Mortality outcomes: all cause (2,775), circulatory disease (1,274) and

cancer (603) deaths (based on ICD-10)

Structural measures of social relationships: Social Integration / Isolation

Social network index (SNI): o Marital statuso Frequency of contacts with friends and relativeso Religious attendanceo Membership in social organization (volunteering)

Biomarkers of Inflammation: hsCRP, fibrinogen, albumin

Data and Measures

UNC Health Registry / Cancer Survivorship Cohort (HR/CSC) 2010 – present; age = 21+; N = 1,105 Mortality outcomes: 114 deaths

Functional measures of social relationships: Social Support

Family; friends Satisfaction with Support

Biomarkers of Inflammation: hsCRP hs 4-plex: IL-6, TNF-α, VEGF, INF-γ

Figure 9. Social Isolation Kills

1.84***

2.03***

1.78***

2.08***

1.62*

1.91*

1.69***

1.88***

1.66***

1.97***

1.44

1.78

1

1.5

2

2.5

3

All Ages 65+ All Ages 65+ All Ages 40-64

All Cause Circulatory Diseases Cancer

Haz

ard

Rat

ios [

95%

CI]

Males

Not Adjusted for Inflammation Adjusted for Inflammation

Note: *p

Figure 10. Social Support Improves Cancer Survival

Models control for age, sex, race/ethnicity;Data source: HR/CSC.

Figure 11. Social Isolation and Inflammation in Adults with Cancer (NHANES III; N = 1,075)

Note: Adjusted for age, sex, and race; results similar after adjustment of other factors.

Figure 12. Satisfaction with Support and Inflammation in Cancer Survivors (HR/CSC)

Note: Models adjusted for age, sex, and race.

3. Social Disparities in Obesity and Inflammation in Young Adults: Behavioral Mechanisms

3 . 1 Y a n g e t a l . A m . J . P r e v e n t i v e M e d i c i n e : u n d e r r e v i e w

Social and Biological Longitudinal Data in Add HealthAdolescence Adulthood

Wave I-II Wave III Wave IV Wave V(12-20) (18-26) (24-32) (32-42)

Social environmental data:school college college workfamily family family familyromantic rel romantic rel romantic rel romantic relneighborhood neighborhood neighborhood neighborhoodcommunity community community communitypeer peer

Biological data:Biological resemblance to siblings in household on 3,000 pairsheight height ht, wt, waist, BMI ht, wt, waist,BMIweight weight, BMI BP, pulse BP, pulseBMI STI test results immune immune

HIV test results inflammation inflammationDNA diabetes diabetes

DNA kidney diseaseGWAS

Harris (2013). The Add Health Study: Design and Accomplishments. CPC, UNC-CH.

Figure 13. Obesity Prevalence in Young Adults by Sex, Race, and SES (N = 7,889)

Note: Wald test for equality of coefficients calculated using logistic regression models adjusting for age, sex, race, and SES; two-tailed.SES Disadvantage Index comprises items for parents’ status at wave I including parental welfare receipt, education and/or income in the bottom quartile of the sample, parent unemployment, and single-parent household structure and ranges from 0 (no disadvantage) to 5 (most disadvantage)

Figure 14. Inflammation in Young Adults by Sex, Race, and SES (N = 6,747)

Note: Wald test for equality of coefficients calculated using logistic regression models adjusting for age, sex, race, and SES; two-tailed.SES Disadvantage Index comprises items for parents’ status at wave I including parental welfare receipt, education and/or income in the bottom quartile of the sample, parent unemployment, and single-parent household structure and ranges from 0 (no disadvantage) to 5 (most disadvantage)

Figure 15. Sex and SES Differentials in Risky Psychosocial and Health Behavior

PresenterPresentation NotesMales had higher levels of social isolation, cigarette smoking, fast food consumption, alcohol abuse, and drug use, but lower rates of physical inactivity than females. The sex gaps declined among those with lower adolescent SES disadvantage.

Figure 16. Race Differentials in Risky Psychosocial and Health Behavior: Males

Note: Wald test for equality of coefficients calculated using logistic regression models adjusting for age, sex, race, and SES; two-tailed

Table 1. Estimated Associations of Social Status and Health Behaviors with Biomarkers of Cancer Risk

Note: +p

Summary

Studies of the biosocial linkages shows how social status “gets under the skin” to influence cancer biology

Social cultural changes occurring in the U.S. young adults continue to shape and modify the projections of cancer burden on the aging society in the future.

A life course approach to cancer disparities helps to illuminate points of intervention in early life periods Adolescent SES and social isolation have lasting influences on

inflammatory and metabolic risk factors for cancer in adulthood Behavioral risk factors are contemporaneous in their associations

with biological risks in adulthood

Acknowledgement

LCCC University Cancer Research Funds (UCRF) LCCC Development Award 2015-16 NICHD P01-HD31921 Carolina Population Center NICHD-NRSA training grant 5-

T32-HD07168

Social Disparities in Biological Risk Factors for Cancer in Young Adulthood: �Obesity, Inflammation, and Socio-behavioral MechanismsOverviewTemporal Dynamics and Cohort AnalysisFigure 1. Estimated Cohort Effects of Cancer Mortality by Sex and RaceFigure 2. Major Behavioral Risk Factors: Cohort Patterns of SmokingFigure 3. Estimated Cohort Effects of Colorectal Cancer IncidenceFigure 4. Predicted Incidence of Colon Cancer by Age, 2010-2030Figure 5. Major Behavioral Risk Factors: Cohort Effects of ObesityFigure 6. Major Behavioral Risk Factors: Cohort Patterns of DietA General Life Course Model of Social Disparities in Health1. Early-life Social Status and Adulthood Inflammation and Metabolic DisorderSlide Number 12Datasets across the Life SpanKey Measures: Harmonized across DatasetsFigure 7. Early-life SES Associated with Biomarkers in Young Adulthood (Add Health): Accumulation of RisksFigure 8. SES Mobility Associated with Biomarkers across the Life Span2. Social Relationships, Inflammation, and Cancer SurvivalSocial Isolation and Tumor Incidence: �Model AnimalsData and MeasuresData and MeasuresFigure 9. Social Isolation KillsFigure 10. Social Support Improves Cancer SurvivalSlide Number 23Figure 11. Social Isolation and Inflammation in Adults with Cancer (NHANES III; N = 1,075)Figure 12. Satisfaction with Support and Inflammation in Cancer Survivors (HR/CSC)3. Social Disparities in Obesity and Inflammation in Young Adults: Behavioral Mechanisms Social and Biological Longitudinal Data in Add HealthFigure 13. Obesity Prevalence in Young Adults by Sex, Race, and SES (N = 7,889)Figure 14. Inflammation in Young Adults by Sex, Race, and SES (N = 6,747)Figure 15. Sex and SES Differentials in Risky Psychosocial and Health BehaviorFigure 16. Race Differentials in Risky Psychosocial and Health Behavior: MalesTable 1. Estimated Associations of Social Status and Health Behaviors with Biomarkers of Cancer RiskSummaryAcknowledgement