Society, Human Development, and Public Health (SHH278)

Society, Human Development, and Public Health (SHH278)

Karestan Koenen Stephen GilmanMarie McCormick

Lisa Berkman

Cardiovascular disease from a life-course perspective

What are the major risk factors for heart disease?

1. Smoking2. High blood pressure3. High blood cholesterol4. Overweight and obesity5. Physical inactivity6. Diabetes

Your guide to a healthy heartNational Institutes of HealthDecember 2005

BMI and Heart Disease

What predicts BMI?

Lawlor et al., 2006, Int J Obesity

N=5,467, Aberdeen birth cohort

Education and BMI

Lawlor et al., 2006, Int J Obesity

Childhood IQ and BMI

Lawlor et al., 2006, Int J Obesity

Childhood IQ and BMI

Controlling for childhood factors

Also controlling for education

No IQ effect on BMI independentfrom education

Childhood IQ and Adult Obesity

Controlling for education, no IQ effect

Lawlor et al., 2006, Int J Obesity

IQEducationBMI

No effect of education on BMI within siblings: why?

Effect of education on BMIIndependent of childhood IQ

When does CVD risk begin?

Socioeconomic status at birth and cardiovascular disease

• Father’s social class and cardiovascular disease mortality: 1.58 (1.32, 1.89) (Smith et al 1997)

• Social class in childhood and coronary heart disease: 1.57 (1.06, 2.30) (Power et al, 2005)

What are the major risk factors for heart disease?

1. Smoking2. High blood pressure3. High blood cholesterol4. Overweight and obesity5. Physical inactivity6. Diabetes

Your guide to a healthy heartNational Institutes of HealthDecember 2005

SES at birth: smoking

Childhood SES1st Cigarette

Regular Use

Quit

Maternal education (4yrs)

1.3

Low parental occupation

1.5

Family poverty 1.4 0.7

Adult SES

Education (4yrs) 1.5 2.5

Occupation

Gilman et al, 2003

SES at birth: blood pressure

Kivimaki et al, 2006

SES at birth: cholesterol

• Father’s social class and cholesterol in adulthood: -0.05 (-.08, -.02) (Blane et al, 1996)

SES at birth: overweight

Poulton et al, 2002

SES at birth: physical inactivity

• Low childhood social class and physical inactivity in adulthood: 1.29 (1.07, 1.56) (Regidor et al, 2004)

SES at birth: diabetes

• Childhood socioeconomic status and mortality from diabetes: 1.52 (1.32, 1.75) (Lawlor et al, 2006)

“Five of six studies reported associations between low early-life SES and little or no adult leisure-time physical activity, five of five found associations with high adult alcohol intake, and eight of twelve studies reported associations with higher smoking rates. Nine of 11 studies reported associations between lower early-life SES and elevated BMI and WHR.”

“Rapidly accumulating evidence is revealing an impact of childhood experiences on subsequent health, well-being, and competence which is more diverse, profound, and longlasting than was ever understood in the past.”

Hertzman & Weins, 1996

Organization of the Course

• Part I: Introduction to life-course epidemiology– Human evolution– Prenatal factors and adult disease– Mental illness

• Part II: Methods of life-course epidemiology– Lessons from clinical trials– Cohort studies

• Part III: Case studies in life-course epidemiology– Health disparities– Reproductive health– Mental health

• Part IV: Focus on specific developmetnal periods Leasons from clinical trials

– Childhood– Adolescence– Older adults/aging

• Course paper and presentation

Public health implications of the early life origins of disease…

• For etiologic models• “Life course epidemiology”• For intervention and prevention

Life course approach to epidemiology

• Life course models: how risk factors matter over time– Latency effects, critical and sensitive

period– Pathways, chains of risk– Cumulative risks

• Dimensions of time in life course research– Disease history– Life history– Social context

Definitions of ‘life course’ epidemiology

• what it is• …”the study of the long-

term effects on chronic disease risk of physical and social exposures during gestation, childhood, adolescence, young adulthood and later adult life. It includes studies of the biological, behavioral and psychosocial pathways that operate across and individual’s life course, as well as across generations, to influence the development of chronic diseases.”

• what it isn’t• “Merely the collection of

exposure data across the life course is not synonymous with a life course model of disease causation.”

• (Ben-Shlomo and Kuh, 2002)

Definitions of ‘life course’ epidemiology

• What– Population-based– Draws on multiple disciplinary perspectives– Elevates the dimension of time – Concerned with dynamic interactions between person and

environment • Why

– Static approaches of ‘modern epidemiology’ (case-control studies, for example) may obscure etiologic processes and intervention points

– Disease risk begins early in life and accumulates over time– Trajectories of risk initiated as early as conception

• How– Design: study persons over time across settings– Measures: developmentally sensitive and appropriate– Analysis: trajectories, timing

More on life course research1. What does ‘life course’ research do?

• From a developmental, sociological perspective, it “broadens the study of individual careers to explicitly locate intraindividual change within a broader historical and social context. Integral to the life course perspective are issues of…• Location in time (history) and place (society and

culture)• Linked lives, integration of individuals’ lives at the

social and institutional levels• Human agency• Timing of lives” (Elder)

2. What does a ‘life course’ orientation offer?• A theory• Framework or perspective• A method

In epidemiology, life course models posit different types of risk factor associations over time, which presumably suggest different causal paths to diseaseEarly-Life Exposures and Risk for Adult Outcomes

• Latency model

• Pathway models– Accumulation

– MediationPower and Hertzman 1997; Power 1991; Lynch, Kaplan, and Shema 1997; Hertzman and Weins 1996

Childhood Adulthood

Latency model

• Typical scenario– Exposure early in life– Disease onset after many years– Risk persists independently of intervening

events

• Application– Fetal programming (Barker)– Critical or sensitive periods– Social and biological risk factors

Evidence for Latent Risks?

• Prenatal famine and adult affective disorder

• Birth complications and psychosis

Brown et al, 2000

Relative risk of adult affective disorder among offspring of mothers exposed to famine in the prenatal period:

1st Trimester 1.23 (0.79-1.93)

2nd Trimester 1.40 (0.98-1.99)

3rd Trimester 1.50 (1.05-2.02)*

Unexposed 1.00

Relative risk of adult psychotic disorders among adults whose mothers experienced hypoxic-ischemia related birth complications: 2.06 (1.93-2.47)*

Zornberg et al 2000

Social & Biological Latent Risks

Relative Risk of Schizophrenia

FAMILY HISTORY

Mother & Father Affected 46.90 (17.56-125.26)*

Neither affected(Population Attributable Risk: 3.8%)

1.00

PLACE OF BIRTH

Capital 2.40 (2.13-2.70)*

Suburb of capital 1.62 (1.37-1.90)*

Rural area(Population Attributable Risk: 34.6%)

1.00

Mortensen et al, 1999

Pathway Models

• Typical scenarios– Risk factors early in life– Cause a series of intervening events– Interact with social and biological factors

throughout the life course– Ultimately lead to adverse health

outcomes

Evidence for latency/pathways effects over the life course

1. Does childhood class predict adult health?– If yes: possible latency effects

2. Does childhood class predict adult risk factors?– If yes: possible pathway effects

Evidence for latency/pathways effects over the life course

• Does childhood class predict adult cardiovascular risk?– RR for low vs. high = 2.6 95% CI(1.5,4.5)

• Does childhood class predict adult risk factors?– Correlation bt adult and childhood SES=.32, p<.001

• Does adult SES predict adult health?– RR for low vs. high = 1.6 95% CI(1.1,2.5)

• Is childhood class or adult class a more important determinant of adult health?– RR for Childhood adjusted for Adult = 2.3 – RR for Adult adjusted for Childhood = 1.4

• Is childhood class an important determinant of adult health after adjusting for sex, family liability, child/adolescent health characteristics, childhood IQ, childhood maltreatment, adult SES:– RR = 1.6 95% CI (0.9, 2.9)

Pathways

• Debate: conceptualizing a causal model of adult illness– Childhood versus adult SES– Causes versus confounders– Mediators versus confounders

“Living conditions cannot simply be left to fluctuate as people pass through childhood and their reproductive and working years and into old age, because health and quality of life at any one stage is affected by prior circumstances and events. …

“The life course may be regarded as combining biological and social elements which interact with each other. Individuals’ biological development takes place within a social context which structures their life chances, so that advantages and disadvantages tend to cluster cross sectionally and accumulate longitudinally.”

Bartley et al, 1997

SHDH Approach to Life Course Research

• Society: Social context– Social– Psychological– Biological

• Human Development: Life History– Development– Life transitions

• Health: Disease History– The progression of disease over the life course– Precursors, prodromes, onset, course, recovery– Continuities, discontinuities

DISEASE HISTORY SOCIAL CONTEXT LIFE HISTORY

Conception,Birth Development

Transitions ToAdulthood Old AgeMidlife

ExposureAccumulation of Risk Factors

LatencyPeriod

Intervening Factors

Early Manifestations(Prodromal Period)

Onset Recurrence,Remission

Socioeconomic Status, Traumatic Exposures, Life Events, Social Connections

Sub-populations: sex, race/ethnicity

DIMENSIONS OF LIFE COURSE EPIDEMIOLOGYDIMENSIONS OF LIFE COURSE EPIDEMIOLOGY

Life course research: it’s about time

• Life course research is inherently developmental in nature• Development = change• We want to observe (and if possible, intervene in) the

processes of change• Locate individual change within the contexts of place,

time, and social hierarchy• Risks for disease often occur early in life

– Origins of social inequalities begin early– Can they be modified later?

• Variations in social context occur throughout the life span– Measures of SES used in health studies may vary

according to the stage of the life span

Policy Implications

• Intervening early in life– Prevent, reduce exposures with latent effects– Avoid negative chains of events

• Interrupting pathways– Prevent (reduce the likelihood of) specific

negative health outcomes– Is it possible to entirely overcome the health

consequences of social adversity early in life?• Multiple adverse health consequences in

adulthood• Multiple pathways