Roy & Moser Jones 1

aboriginalconspiracyUrban and Civil

Nov 16, 2013 (3 years and 8 months ago)

144 views

Roy & Moser Jones
1











Integration of family science and public health

through the life course framework


Kevin Roy, Ph.D.

Marian Moser Jones, Ph.D.
, M.P.H.

Department of Family Science

School of
Public Health

University of Maryland, College Park




In this paper, we first will explore the state of the life course framework in family science
research and public health research separately. We will identify how these frameworks overlap
in a growing common ground and shared understanding of change over
time, and provide a
multidimensional model with three “key moves” that will pull family science and public health
closer together in utilization of life course concepts. Finally, we discuss five opportunities and
challenges within an integrated life cours
e framework, each of which involves tradeoffs and risks
that will shape the pursuit of a shared vision for families and health research and intervention.






Roy & Moser Jones
2


Across the United States, more and more Family Science Depa
rtments are moving
towards
for
mal or

informal school collaborations under t
he umbrella of health
. Many factors
drive

this movement: the need for resources and funding;
the availability of innovative
longitudinal secondary datasets (Elder & Taylor, 2009; Hauser, 2009);
the logic of a broad
conceptualization of health across the soc
ial sciences; and the possibility

of
coordinated
interventions

through public health

institution
s. The
fields o
f family science and public health
also have weaknesses that can be addressed through new partne
rships
. In family science,
researchers
seldom address
specific health

concerns

in family
and larger kin
networks
, such as
chronic illness, asthma, or smoking
. In public health, researchers and practitioners
strive to
understand how family
processes

shape healt
h outcomes

and to

see families as sites for
intervention
.

Additionally, public health researchers

view
of
health

may be grounded in
a
macro
-
population perspective that focuses on large
-
scale

mobilization

in communities to prevent
disease, wher
eas many family science researchers’ view of health emerges from the
micro
-
context
of family dynamics
.
Integrating the perspectives
of the two fields can prove fruitful.


The two authors of this paper come together with different backgrounds but an inter
est in
building common ground. Roy was trained in human development and social policy, and has
cond
ucted research on the life course of low
-
income fathers

over the past 15 years

(
Marsiglio &
Roy,
2012
;
Roy, 2006; 2008)
. He has begun to explore contexts o
f health of young men,
especially depression, trauma and loss, and family care, in recent studies. Jones
earned a
masters in public health degree a
s well as a doctoral degree in s
ociomedical science, with a focus
on the history of public health.
Her
work is focused on development of human services over
historical time, as well as interaction of biography and history

(Jones
, 2013
)
.


Roy & Moser Jones
3


Why choose life course as a bridge between the two fields? This paper is not an attempt
to compare possible theoretical
linkages. Both fields have numerous grand and mid
-
range
theories that are specific to them. The bioecological model and the social determinants models
of health model are quite similar and provide both fields with common ground in thinking about
the valu
e of context and multiple levels of analysis. However, life course frameworks have
begun to emerge in both fields, independently, and they carry over an attention to context, but
coupled with a critical focus on change and development over time.

A paradig
m that is rooted in dynamic change over time speaks to our core understandings
of what families are and what health is. Families are the way that many individuals experience
time: through the celebration of family birthdays, through the length of a marri
age, through
“feeling old” when one’s chil
dren leave home. Health

is probably best described as a dynamic
arc,
a long
-
term pathway of growth and decline over many years
. The World Health
Organization’s broad definition of health (“complete physical, ment
al, and social well
-
being, and
not merely the absence of disease or infirmity”)
attunes us

to
health
contexts that shift over time,
to precursors and their
much later
consequences.
Such a paradigm can
identify

opportunities for
intervention in an earlier
time period to change life trajectories and improve health over the long
-
term.
As AJPH editor Stella Yu stated in a 2006 editorial,
“By systematically pursuing the life
-
course paradigm, we could potentially reduce the heavy human and economic costs prec
ipi
tated
by health inequities
.


However, before jumping onto this bridge, it is important to examine
whether the two disciplines are talking about the same thing when they t
alk about the “life
course”
framework
.

In this paper, we first will explore the state of the life course framework in family science
research and public health research separately. We will identify
how these frameworks overlap
Roy & Moser Jones
4


in a growing common ground and shared understanding of change over
time, and provide a
multidimensional model with three “key moves” that will pull family science and public health
closer together in utilization of life course concepts.
Finally, we discuss
five opportunities and
challenges within
an int
egrated life cours
e framework, each of which involves tradeoffs and risks
that will shape the pursuit of a
shared vision for

families and health

research and intervention
.



Life course framework in family science research


The life course perspective in family science offers tools for understanding the dynamics
of relationship and identity formation in context and over time. Life course theorists consider
ways in which individual lives both shape and are shaped by social s
tructure (Bengtson & Allen,
1993)
,

examining

how changing social contexts transform normative roles

in families, such as
being a mother, father, grandparent, or sibling.
The framework has its roots in sociology and
comparison of multiple cohorts
, but over the past two decades it has been utilized to examine the
complex interplay between cohorts and generations as they are shaped by social changes.
Elder’s research on the children of the Great Depression (1999) is perhaps the best known life
cour
se study of family life, with Laub and Sampson’s multiple
-
method study of desistance from
crime (2003) serving as a more
recent

model.

In general, a

life course perspective
in family
science
offers four

sensitizing concepts that frame relationship

and ide
ntity change over time:
human agency, linked lives, location in context, and multiple rhythms of time (Giele & Elder,
1998).

Researchers have utilized multiple methodologies with these concepts, including
longitudinal ethnography (Burton, Purvis, & Garre
tt
-
Peters, 2009).

In the field of fa
mily science,

different strands

of theoretical frameworks
compete for
what is considered to be the most effective theory of development and change for individuals and
families

(White & Klein, 2008)
. The life course fram
ework noted above stands in comparison to
Roy & Moser Jones
5


the life span approach,
which is
primarily based in psychological research. Th
is

emphasis in life
span work is similar to th
e perspective embraced in
the health

sciences: an assumption that the
unit of analysis is the individual’s ontogenetic developmental pathway, and that a normative
pathway to optimal well
-
being can be studied and predicted.
A unique offering from family
science is the family development fram
ework, which has enjoyed renewed interest in
recent
years. In contrast

to
other

approaches
, the unit of analysis is a family unit, with an assumption
that most families progress through common stages of development, such as childrearing, empty
nest, or re
tirement. To link commonalities across these thr
ee different but related frameworks
,
James White suggested a transition theory (2005), with a focus on common patterns of family
role transition and change over historical, family, and individual time.


T
he
promise of the life course perspective has not been realized in the study of health by
family scientists.
Study designs are often limited to examining the effects of family structure on

generalized health outcomes. In a recent decade review of literature

on families and health, Carr
and Springer (2012) indicate that the emerging challenge is to explore
“under what conditions,
for which outcomes, for whom, and through which pathways do family structure, context, and
process af
fect health?” In other words,

there is a critical need for a paradigm that
brings

processes, context, and change over time

into the analysis of family health
.


The
l
ife course

paradigm

in public health


Over the past

fifteen years
,
t
he life course perspective has
become increasingly
influential
in

public health

research, and has spilled over to policy and programming

(
Ben
-
S
hlomo, 2007;
Davey Smith, 2007
).

The core mission of public health focuses on health promotion and disease
prevention at the community and population level (IOM, 19
88).
If
“most public health programs
Roy & Moser Jones
6


today are dedicated to remedying the damage p
receded by earlier deprivations
;”

the life course
perspective offers a
new paradigm for
address
ing

these
distal
causes of disease.
(Yu, 2006)
.

Public

health researchers have

begun to

embrace the life course perspective
because,
unlike traditional models of disease causation and prevention, it offers a mechanism to account
for temporal aspects. Specifically, the life course framework enables researchers to model the
long time
lag between exposures to disease
-
causing or health
-
protective agents, and the health
outcomes pro
duced by these exposures. It also

integrate
s

mediating factors between the initial
exposure and the ev
entual health outcome into time lag

models

(Ben
-
Shlomo a
nd Ku, 2002)
.
This modeling is particularly relevant when considering social and environmental causes of
disease such as
toxic chemicals, second
-
hand smoke, or low socioeconomic status

that may not
have an immediate impact on the health of populations.

Add
itionally,
the
positive health
factors
that may protect populations from
chronic disease
often play out over decades, not days
,
weeks
,
or months
. The life course framework

has proved particularly useful in modeling the influence
of
prenatal or early life
expo
sures on later health outcomes, and in thus strengthening the case for
health policy and programming that focuses on providing intensive services at
critical
periods
in

early development rather than on health care later on in life (Richardson
et al.
, 2
012).

Finally,
the life course approach
offers
a means to address one of the most prominent
concerns among U.S. public health researchers

health disparities.

Researchers looking to
locate the sources of these disparities in immediate biomedical factors or

social context have
been frustrated. For example, the cause of the twofold disparity in infant mortality rates between
African Americans and whites in the U.S. does not appear to be attributable to the mother’s
current
socioeconomic status,
prenatal
healt
h care, or education level, but may lie in factors
related to the mother’s own childhood
environment (Lu
et al.
,
2003,
2010)
.

Leaders in the public
Roy & Moser Jones
7


health field have
highlighted
the

life course perspective as a way to explain these gaps

and

t
o
close them:

“The knowledge gained from

life
-
course studies could
be resolutely applied to health
and other programs in different age, racial/ethnic, socioeconomic, and gender groups to relieve
suffering and offer hope of livi
ng healthy and fulfilling lives
”(Yu, 2006)
.



Tentative (and growing) c
ommon gro
und

The

concurrent embrace of t
he life course approach in
family science and public health
may seem to open up a convenient bridge between the two disciplines
.

But

there are important
semantic and
conceptual
differences between the two life

course paradigms that
may

interfere
with

such cooperation if
left
unaddressed.
Unlike
the social science framework most often used
in family science research
,
the

life course

approach most
commonly employed by public health

researchers

is a stripped
-
down model
. E
arly
-
life

inputs are generally

limited to
factors that can
be operationalized as measurable variables

in large datasets

(i.e.,
maternal

body mass index
(BMI)
, prenatal doctor visits
,
parental income/ education/socioeconomic status, maternal
depression as measured by Beck Depression Inventory
(BDI)
)
,

and the later
-
life outcomes are
narrowly constructed as

quantifiable
dependent variables such
as
BMI
, asthma, or coronary heart
disease (C
HD)
, or individually diagnosed mental illnesses
.
As a result of this heavily
-
circumscribed approach, much public health research tends to
exclude complex systems such as
families
from the
analysis

of the relationships between exposures earlier in life and
health comes
later on
.


A foc
us on processes


like those

within families
-

is sacrificed
for the goal

of
precision.

What is identified as a “health outcome” may
also
differ substantially in both fields.
Family scientists, drawing primarily on social and
behavioral science, typically examine g
eneral
Roy & Moser Jones
8


health outcomes. T
opics might include
addiction and substance abuse (Kerr, 2011),
mental
health
issues
such as depression or stress (Wickram, 2012; Higgins, 2010
; Turney, 2011
),
intimate partner violence (Wie
rsma, 2010; Slep, 2011),
suicide (Denney, 2010), or general
physical and mental health (Schmeer, 2011). In contrast, public health researchers commonly
pursue specific health outcomes mor
e in line with epidemiology, maternal and child health, and
e
nviron
mental health science, including asthma,

low birth weight,

obesity, smoking, diabetes,
lead poisoning, and related diseases

(see Bauldry
et al
.,2012; Epstein,
et al.
, 2012)
. In effect,
researchers from both fields address health outcomes, but diverge on
type and specificity.

Fig. 1 Multidimensional Model of Life Course Integration in Family Science and Public Health



Processes within families



Family processes in context


Epidemiology







Social and behavioral science







Health
administration/policy







Environment







An integrated life course framework must work within, and across, all of the core
knowledge areas that define public health training. For the sake of this theoretical paper, we
have condensed them into four areas: epidemiology; social and behavioral scie
nces; health
services administration; and environmental health sciences. (Note: biostatistics is excluded due
to its exclusive focus on methods). Similarly, the framework must address the range of family
processes that occur within families (and househo
lds), and outside of households with family
members in local and global communities. Figure 1 illustrates a multidimensional model of key
“move” #1

“move” #2

“move” #3

Roy & Moser Jones
9


overlaps between family science and public health. Each of the eight cells represents active or
potential research i
n overlapping dimensions of the two fields.

We identify three critical “moves” that would
pull

family science and public health closer
through the use of a life course framework.
The first “move”

is to mo
re closely link
epidemiological and social &
behavioral studies together,
paying attention to processes within
families and to family processes in context. This linkage can be facilitated by understanding the
similarities between life course models in epidemiology and in concepts in family science
(as
advanced by Elder).

Within the broad field of public health research, the life
cour
se approach

has been most
actively pursued within
epidemiology
.

Epidemiology, “the study of the occurrence and
distribution of diseases and other health
-
related conditio
ns in populations” (Kelsey, Whittemore,
et al.
, 1996),
rose to prominence as a
public health discipline
in the
1950s
,
when
seminal
epidemiological studies definitively established
the causal relationship between smoking, CHD
and lung cancer
. Other
important studies linked diet

and
hypertension to CHD a
nd cancer
(
Doll,
Hill, 1950, 1954, 1956, 1964
)
.

These studies

appeared just as science was appearing to triumph
over epidemic diseases such as polio

and

tuberculosis,
and
as
medicine paid increasing a
ttention
to chronic diseases such
as
cancer and CHD. By establishing the key risk factors for these late
-
twentieth century killers, this research
enabled
public health

policymakers and

practitioners
to
target
individual behaviors

i.e. smoking, high
-
animal
fat diet, inactivity
--
in
health promotion
and disease prevention campaigns.
However, by the 1980s
epidemiologists had
beg
u
n to struggle
with the
limits of
th
is model
:

the classic
CHD
risk factors
(
for example,

including high
cholesterol, hypertension, smoking, and physical inactivity
)

could
only explain 75% of th
e cases

(Beaglehole, Magnus 2002)
.

Additionally,
regardless of

individual

health behaviors
,

major
Roy & Moser Jones
10


differences in the health of populations and communit
ies remained. These differences


termed
health disparities when applied to racial and socioeconomic groups

appeared to be

linked to
socioeconomic status,
environmental exposures
, and other larger
-
scale factors, but traditional
epidemiology did n
ot have th
e tools to explain these relationships well
.

This

shortcoming led
epidemiologists to adopt
other paradigms including the life course

approach
.


Life Course Epidemiology
, first named as such in a 1997
monograph

(Kuh and Ben
-
Schlomo, 1997),

has transformed the life course paradigm into a testable set of hypotheses that
researchers can use to interrogate specific associations between exposures and health. These
hypotheses

generally
posit a

causal
association between exposures

(E)

in population
s and health
outcomes (O) that occur years later.

These hypotheses can be
divided into three principal

models
(Lynch and Davey Smith,
2005, Giles
et al.
, 2011)
.
The first
,
the
“critical period
” model
,
posits that
the timing of the
exposure
(
s
)

is what matters most
, and that early exposures can determine later disease status
regardless of later intervening factors

(Lynch and Davey Smith, 2005
,
Kim &

Richardson

et al.
,
2012
).

The second “accumulation” hypothesis focuses on multiple exposures tha
t add up over
time. The third hypothesis, the “pathways” model, posits that the sequence of exposures and the
trajectory followed may exert significant effects on the outcome
. Disagreement remains on
whether these hypotheses can be disaggregated from one a
nother, and on how to name the
m.

(Giles
et al.
, 2011;

Hallqvist
et al.
,

2004).

Nevertheless researchers over the past decade have
shown increasing interest in using these life course models to ask big questions about life events,
health, and disease.

The c
ritical period hypothesis conceptually reflects

Elder’s
life stage principle


that the
“personal impact of any change depends on where people are in their lives at the time of the
Roy & Moser Jones
11


change


(Elder, 1994).

Examples of “critical periods” of exposure

that epidemiologists have
employed in such analyses

include
short windows in fetal development
where an infection or
exposure to drugs

(such as thalidomide)
can have devastating consequences

on subsequent
prenatal
development
,
whereas t
hey might have limi
ted consequences if occurring

a week or a
month before or after this window
.
This model overlaps significantly with the Developmental
Origins of Health and Disease (DOHaD) theory, which holds that fetal programming, especially
undernutrition during gestati
on, permanently shapes the body’s metabolism, structure and
function, and greatly influences a person’s propensity to develop cardiac and metabolic disorders
later in life (Wadhwa
et al.
, 2009).
These examples emphasize the persistence of such prenatal
eff
ects in the face of later life events, rather than the ability of later life choices to modify or
reverse them.
Critical period models can also

look at “sensitive periods” of childhood
development such a
s school entry or early puberty
, when the effect of
an exposure is greater that
it would be in another period

but may still be reversible
or modifiable later on. T
he primary
em
phasis in this hypotheses are critically
-
timed exposures and not

later modifiers.

In the accumulation hypothesis, m
embers of a cohor
t
are exposed to a variety of negative
exposures


lack of prenatal care, second
-
hand smoke, pollution, poor diet


and these factors
accumulate
, like debts,

to produce more h
ealth problems later on in life. Positive exposures such
as prenatal and perinatal

care, early childhood education,
or
a healthy neighborhood add up to
create a positive health “bank balance” or help to pay off the “health debts
.


What matters

most

is the total “health bank balance,” not how the person arrived at this figure.

By
contrast
,

the third hypothesis,
or
the
“pathways”
model
, posits that the sequence of
exposures and the trajectory followed may exert significant effects on the outcome. In this way,
the pathways model transforms Elder’s idea of “transitions and trajectorie
s” into
a
testabl
e
Roy & Moser Jones
12


hypothesis
.

I
n some versions of the pathways model both the timin
g and kinds of exposures
matter

(This model, also called an effect modification hypothesis, can thus incorporate the
critical time hypothesis).
C
hains of causality operate,

in which one exposure

say
,

lack of
physical activity in late childhood


leads to another


overweight in early adolescenc
e


which
can lead to a later
health outcome such as Type 2 diabetes
or
heart disease in
adulthood. The
chains of causality
can be
mor
e complex than this

example illustrates
. They do not have to
proceed in a simple chain (E(1) >> E(2)>>E(3) >>> O) but can involve multiple pathways of
causality (i.e., (E(1) +E(2) >>E(3) >>O)
and can
incorporate later life experiences that magnify

or m
oder
ate

effects
of
early
-
life exposures

(Kuh
et al.
, 2003).

Importantly, these life course hypotheses do not pay much attention to some of the
“fuzzier” concepts that lend richness to Elder’s framework.
H
uman agency
, a key notion as
individuals adapt in
periods of social change,

is rejected or deemphasized. This is not an
omission:
the life course framework represents a response to the limits of the risk factor
paradigm, which assumed that individual health behaviors
, and thus the lifestyle choices that
i
ndividuals make,

constituted
a key
determinant of health. A critique of the idea that “you are
responsible for your own health” was sorely needed, and
life course epidemiology is one model
that supplies it.


L
ife course epidemiology
also has not focused mu
ch on
historical time/context and
interlinked lives
, two other main aspects of the social science life course perspective
.


In these
cases, the omission is not necessarily intentional. Historical time could
be incorporated in the
“critical time” hypothesis
,

as a critical exposure

could

result from a
specific historical event such
as a famine, a natural disaster,
a sudden economic collapse causing a food shortage,
or

a war that
causes

people


including pregnant women


to become refugees
. Similarly, histori
cal events
Roy & Moser Jones
13


could be incorporated into chains and pathways of causation. The notion of interlinked lives can
also be incorporated into these epidemiological models, in that maternal, paternal, and family
-
level influences could be included as exposures in

ca
usal pathways, and adult family structures
or measures of familial well
-
being could be measured as outcomes.

Thus far, however,
life course epidemiology has
generally excluded family
-
level factors
either as exposures or outcomes. It has
focused on individu
al
-
level (physical activity, diet, health
care), neighborhood
-
level (availability of recreation facilities), or population
-
level factors (SES)
as exposures (Deardorff, 2012, Lynch 1997).
S
pecific maternal health factors such as
prepregnancy
,

BMI (Fleten

et

al.
, 2012)
,
maternal depression (Giles
et al.
, 2011)
,

or perinatal
interventions such as infant feeding (Owen

et al.
, 2005) or visiting nursing care (Eckenrode
et
al.
. 2010)
,

have been anal
yzed. Other

studies have also included measures of maternal or paternal
S
ES and family size as variables

(Jokela
et al.
, 2009
).

One
noted exception is a recent French study that

used
a large
-
scale dataset collected

in
the Paris metropolitan area to examine

possible c
ausal associations between the family
environment that respondents reported experiencing as children (<18) and their perceptions of
their own physical and psycholo
gical health in early adulthood

(Roustit,
et al.
, 2011). The
researchers were a
ble to define
the “exposure”
of family relationships using several indicators,
which they were able to quantify as variables in the model: poor maternal relationships, poor
paternal relationships, intraparental violence during childhood, and parental separation and
divo
rce. They found that the quality of family relationships during childhood did significantly
affect self
-
perceived health during early adulthood. More surprisingly, they found that family
violence and poor parental relationships acted as mediators

intermedi
ate factors that explain
and predicted the association between separation and divorce of parents and negative health
Roy & Moser Jones
14


outcomes during adulthood. In other words, parental divorce on its own was not an independent
predictor of poor health in adulthood. This i
s only one study, and it is not conducted on a U.S.
cohort, but it illustrates the possibility of operationalizing family processes within the life course
epidemiologic model to study the longitudinal relationships between family

influenc
es

and
specific
he
alth

outcomes
.

A second “move”

is an infusion of family science research into the traditional focus on
health service administration and, more broadly, health policy as it relates to families. Multiple
domains in both fields have studied the interaction o
f families in their communities, as such
interactions are related to health outcomes. For example, Montoro
-
Rodriguez (2012) examines
school
-
based mental health services for families, and Richardson (
2011
) details the choices of
family members who turn to the criminal justice system to provide rare access to mental health
services for their young adult sons.
Rapidly
-
approaching changes in

the US

health care system
with the advent of the Affordable Care Ac
t present opportunities to study family processes as
they interface with new and emerging systems of care.

Less developed, however, is our insight into processes within families that lead to service
use. For example, adult men are typically less likely
to seek out health services, and many rely
on women in their family networks (wives, mothers, sisters) to arrange these appointments. How
do family members strategize to motivate men to visit their doctors for annual well
-
being
appointments? Are strategi
es different for young men than older men past retirement? Life
course researchers in family science would emphasize the multi
-
generational arrangements that
might drive health literacy efforts and family
-
level processes of health surveillance, all of whi
ch
link to issues of access, availability, and outreach.

These insights have emerged as well among
maternal and child health scientists, whose focus on provision of prenatal care has been
Roy & Moser Jones
15


enhanced by a longitudinal, ecological approach that incorporates e
xtended family members and
community actors to promote the well
-
being of infants and mothers (Pies, Parthasarathy, &
Posner, 2012).

Further, MCH policymakers have

increasingly begun to apply a life course paradigm in
tackling stubborn public health chal
lenges such
as
racial disparities in birth outcomes and the
ranking of the U.S. as worse than 45 other nations in infant mortality rates.
Healthy Start, a
federally funded program that has previously focused more narrowly on providing prenatal care
to high
-
risk pregnant women in low
-
income areas, now has begun to focus in numerous localities
on improving women

s health behaviors and health care throughout the “maternal life course”

before conception occurs, through the pregnancy and
during
critical periods
of infant and child
development

(Fine
et al.
, 2009;

Baltimore Healthy Start, 2012). More broadly, MCH
policymakers have begun to advocate for a shift in
the way that primary healthcare is delivered
to
families and children, from the current fragmented syst
em to one that provides “continuity of
care, starting before pregnancy and continuing throughout life.”
In this new service delivery
model, families will be able to
establish a “medical home”


a primary care provider that
remains
the same regardless of jo
b and income changes, and allows them to

chart a course
for
long
-
term
health


(Fine
et al.
, 2009
).
Long
-
term health factors that increase the likelihood of a healthy
pregnancy in women, such as a
healthy lifestyle,
obesity prevention
, stress management, and
knowing the
family history of genetic conditions, wi
ll be introduced and promoted throughout

childhood and adolescence, rather than just during pregnancy

(
Grason

&
Misra
,
2006)
.

In “
the third move
,” family scientists and public
health researchers confront the
challenges to build new conceptualizations of human and en
vironment interaction.
Public health
researchers have begun to employ the life course

framework to explore how prenatal and early
Roy & Moser Jones
16


childhood
exposures to

a negative s
ocial environment, built environment, and natural
environment affect adult health outcomes

(Glass & McAtee, 2006)
. Studies have examined the
link between these early exposures and

disabilities

in later life

(Bowen and Gonzalez, 2010),
anxiety and
depression

in adolescence and emerging adulthood

(Najman
et al.
, 2010)
,
and
obesity

throughout childhood

(
Trasande
et al.
, 2009).

Research
ers

ha
ve

posited chronic toxic
stress


such as that caused by chronic neglect and abuse, parental substance abuse
,

de
pression,
or violence
--

as a mediator of the relationship between childhood poverty and later adverse
health outcomes, as prior studies have established a relationship between toxic stress early in life
and

premature death or
elevated rates of chronic dis
ease later in

life
. Physical environmental
exposures have been suggested as other mediators:
research has established a strong association
between poverty

and high
blood
lead levels in children ages 1
-
5,

and between high blood lead
levels and
impaired cogn
itive functioning
(Miller
et al.
, 2011), while early exposure to
poll
utants, including
second
-
hand
smoke, has been linked to impaired lung functioning

/asthma/COPD in later life (Wang and Pinkerton, 2008). But research
ers in public health have
only begun t
o examine

families as

buffer
s

or multiplier
s

for the effects of toxic stress or exposure
to environmental toxins
, and
have yet to incorporate family
-
level factors into

life course
epidemiology.

In this area, knowledge is being disseminated unevenly, and
sometimes “backward,”
from practical settings and popular literature to the research arena. Steingraber (2011), for
example, presents a book
-
length examination of looming toxic environmental effects


such as
chemically
-
treated playground equipment


on c
hildren, and how parents monitor and strategize
to deal with such neighborhood threats. Maring and colleagues (2010) describe the Healthy
Roy & Moser Jones
17


Homes Initiative, designed to rehabilitate homes for low
-
income and working poor families who
have faced decades of ne
glect expressed in high rates of asthma and lead poisoning

Part of the reason that family factors have been considered in practice and not in research is that
data has been lacking. But new large
-
scale datasets on children’s health in an environmental
cont
ext, discussed in more detail below, can provide opportunities for such research.

Relationship
processes within families


as environments are brought into household
routines


are in need of greater attention from both fields

as well
.
The effect of paren
tal
smoking on prenatal and children’s hea
lth is well documented, but which

family processes are
supportive to parents quitting smoking, and which ones interfere with smoking cessation efforts?
How do parents’ time management and relationship dynamics aff
ect their ability to provide
consistently healthy meals for their children
? Which families can pro
-
actively reduce use of
plastics, and exposure to BPAs

which have been linked to adverse health outcomes in children
-
-
and which families
lack the access to in
formation and resources to
switch away from plastic?
Family scientists may offer insights into intra
-
familial processes that unfold

over time, across
generations
, to either positively or negatively mediate the effects of environmental toxins and
negative
social environments.


Opportunities and challenges


The promise of an integrated life course framework does pose a number of critical
challenges. These challenges are well
-
understood; they have confronted both fields for decades,
although they take on new

urgency with current policy initiatives
;

with
demographic change as
American families

grow

older, more diverse, and
more complex
;

and with the maturing of social
and health sciences, through use of innovative methods, technologies, and emerging worldviews
.
Roy & Moser Jones
18


In short, the
se challenges
present unique opportunities at the same moment, for rigorous science
and effective intervention
s

to advance the health of families and communities.


Multiple types of change over time

T
he varied approaches across
both fields
needed, as first step, to
conceptualize
change
over time,

and they have done so with similar attention to individual
(or ontogenetic)
development. The notion of a timeline, or looking at lives “from the long view,” compels us to
see how early experiences
influence later health outcomes. Similar to the focus of life span
psychology on the experiences on an individual over time, the insight that early or “upstream”
events have health consequences later in life is
still
fairly revolutionary. We can examine these
types of effects within a stage or critical period of life


such as early childhood or mid
-
life.
This
limited perspective of transitions is challenged, however, by an emphasis on family
-
level
transitions


in m
arriage or child
-
bearing


that are experienced in relationships and
simultaneously shape the health outcomes of multiple actors at once (Carr & Springer, 2012).
The challenge for both fields is that traditional data collection is limited to an individual

re
port
within a family network. For example, h
ow
often do we collect data to
capture divergent health
outcomes for two young spouses who experience a pregnancy termination together?

Linking these important life events and transitions, researchers in both fields have
identified pathways or trajectories of individuals, or of cohorts.
Life course r
esearchers in public
health
are
pay
ing

close attention to critical life span periods or tu
rning points/transitions, such as
pregnancy,
early childhood,
adolescence, and retirement because health interventions during
these critical periods can shape longer
-
term health outcomes. Moreover, Elder (1996) provided
the concepts of timing, sequence, a
nd duration to describe the nature of these transitions in
Roy & Moser Jones
19


biological, psychological, historical, and cultural experiences
that shape

the pathways of
individuals and even populations (Halfon & Holstein, 2002).

One of

the best known example
s

of
research th
at crosses family science and public health is the work of Geronimus and colleagues
(2006), which explores the accumulation of risks that result in health inequalities over the life
course.
Emerging methodological techniques in the life course toolbox all
ow researchers in both
fields to identify mechanisms of cumulative disadvantage (O’Rand, 2009).

Longitudinal e
ffects on individuals can be
also
aggregated to find patterns across birth
cohorts

in the population
.

They can
be seen to emerge from shifting
structural, environmental,
and social changes, through historical or “period” effects. With the emergence of datasets that
feature multiple waves of measures of structural change and individual change, public health and
family science researchers alike ha
ve successfully teased apart age
-
period
-
cohort effects. For
example, Ananth and colleagues (2001) found consistent age effects across periods and cohorts
for pre
-
term delivery rates among White and Black women. They suggested that biological
factors play
ed an important role, although some historical changes, such as the introduction of
medical technology
,

increased the
survival rates of premature infants.


In order to
describe the complexity of family relationships with and across

families,
however, fami
ly scientists
must
distinguish between birth cohorts within a population and
historically
-
bound cohorts within a family, also known as generations
(George, 2009).
The
notion of “generational time” is common in family development theory, but relatively
una
cknowledged in public health research. Only a multidimensional view of time that includes
age, period, and cohort will allow all of these researchers to describe the complex patterns of
family dynamics over the course of births, deaths, and shifts family
membership over many
years
.

Roy & Moser Jones
20


Advantages of nested systems models

B
oth fields of research and practice have embraced
a systems model
, whether linked to
the pioneering work of Bronfenbrenner or to bioecological
/social determinants of health

model
s
.
All of
these visions of determinants at multiple levels are embedded in a life course approach. In
effect, a life course framework “puts wheels on” a multiple level systems model, encouraging us
to examine dynamic changes over time. The life course framework can

t
herefore link the
cohort
level changes in macro pathways with changes in micro pathways in families, dyads, and
individuals.
Pub
lic health research generally treats families

as a “black box”
,
synonymous with
household unit
s
,

or through
reports on family

patterns and correlations with health outcom
es
.
Family science can fill
that black box and offer a more sophisticated underst
anding of
interactions among

family members
and between

families and their social environments and how
these affect health behavi
ors and health outcomes.

There are common threads that emerge from early statements on application of a life
course framework to health development (Halfon & Hochstein, 2002) and more recent guidelines
to shape the practice of maternal and child health (Fi
n
e & Kotelchuck, 2011). P
ublic health
researchers
have in recent years increasingly embraced

framework
s

that
acknowledge multiple,
nested determinants of health,

and
have stressed
that such a framework must account for how
biology and context interact

(Glass & McAtee, 2006)
. Some models of life course in public
health are driven in part by mapping the emergence of inequity, not simply through biological

and individual
-
level behavioral

factors
,

but
through
social
/structural

factors as well (Fine &
Kote
lchuck, 2011).

Likewise, family scientists have begun to integrate biomarkers into their longitudinal data
collection efforts, which may lead to rich insights into how families shape health outcomes of
Roy & Moser Jones
21


their mem
bers. Sibling studies, which have

served as

the common approach to examine gene and
environmental interaction, hold

renewed promise for understanding family health. For example,
Guo, Ro
ettger, and Cai (2008) acknowledged

genetic propensities for adolescent delinquency
among siblings in the same fa
mily. However,
the potential negative effects of
these
polymorphisms wer
e dampened by a family process
-

namely, adolescents sharing daily meals
with their parents.
Such
innovative
research re
flects the

deep potential for theory development
across

publi
c health and family science,
facilitated by data on the interplay of biological and
social factors in families

(D’Onofrio & Lahey, 2012)
.


Individuals, linked lives, and variation

When we note how public health and family science
are

explicitly

challenged to find
shared approaches to address change over time and multiple levels of analysis, there are also
implicit

challenges. Related to framework for change and context, both fields take distinct
positions on locating health “within” the individ
ual.
P
ublic health research
typically does not
recognize families as units of analysis or processes in and of themselves
, except in the most
crude terms, as households or “family units”
.
However,

most

researchers in family science are
themselves not expe
rts in family
-
level analysis. There are few true
“whole family” studies of

all
of the complex relationships as they unfold. H
ess and H
andel’s case studies of five family units
in
Family Worlds

(1959
)
still stand

as a landmark recognition of the difficult
y of addressing the
experiences of each member of a family.

Again, Elder’s concept of linked lives may be a
critical bridging concept as it suggests a
way to think about how change in one person’s life (the onset of a chronic condition, for
example) can ri
pple into the lives of a child, a sibling, or a spouse.
The shifting dynamics
Roy & Moser Jones
22


between spouses over time, as they confront paid and unpaid work in the labor force and family
households, lead to clear gender
-
related health disparities in later life (Moen &

Chermack, 2005).
Moving from dyadic modeling into new levels of complex network relationships, Smith and
Christakis (2008) have demonstrated how interwoven webs of family and friends shape network
members’ health, including smoking behavior, obesity, and

happiness.

In addition, most of the focus on individual health is rooted in an assumption of
normative, linear pathways of development. In this way, the
biomedical model

embraced by
some
public health professionals
can
mirro
r those of developmental
scientists. This model

describe
s

homogenous stages but may not account for the heterogeneity and variation in
communities, families, and individual experiences. The life course framework in family science
does offer tools to better understand variation i
n traject
ories and experiences over time
(Bengtson & Allen, 1993). It moves beyond modal or average
types and addresses deviations
and diversity.
J
ust as individuals differ by gender, ethnicity and race, or SES, there is structural
diversity within famil
ies, which leads to distinct health disparities within and across families
over time.

For example, m
any researchers describe how aging leads to increased heterogeneity in
health outcomes among birth cohorts (Dannefer’s “Matthew effect”).
Some of this he
terogeneity
may be related to family experie
nces of individuals. C
ompared with couples who are
continuously married, men and women who experience marital loss are more likely to develop
cardiovascular disease later in life (Zhang & Hayward, 2006).
It is
challenging to examine
dyadic couple relationships, but even moreso to analyze complex extended kin networks with
multiple members. These kin networks typically exhibit

growin
g heterogeneity

as they add and
lose members through birth and death, marriage a
nd divorce (Bengtson,

et. al., 1990).
In effect,
Roy & Moser Jones
23


a bold move to look closely at processes within families and how they
lead to distinct

and
divers
e
-

patterns of health outcomes may be a risk for researchers who gravitate to models of

normative individu
al

development or population
-
level

patterns.



Advent of large
-
scale datasets

One reason that these life course models have become more attractive to
researchers in
public health and family science

in recent years is that large, robust longitudinal datasets have
only now become available in the U.S, and elsewhere. In the past, most longitudinal cohort
studies (studies looking at a large group of similar people over a long period of time) took place
in the United Kingdom, where scientists have been able to utilize data collected by the National
Health Service. But over the decade or so, numerous U.S. datasets have become available for
study, including the Early Childhood Longitudinal Study, the Nation
al Longitudinal Study of
Youth,
and
the National Longitudinal Study of Adolescent Health. The new National Children’s
study, for which data is now being collected, promises to
greatly
expand possibilities for analysis
in the future (Trasande
et al.
, 2009,
National Children’s Study 2012). Large datasets in other
countries such as France and Australia have also been made available to researchers (Roustit,
et
al.
, 2011, Cameron
et al.
, 2012). Additionally, the increasing availability of statistical software
ca
pable of processing analyses of these large
-
scale datasets, and the development of
sophisticated statistical modeling techniques, is making such
analysis increasingly feasible
(Gamborg, Jensen,
et al.
, 2011).

In the arena of environmental health, these la
rge
-
scale datasets have yet to be fully
utilized, as the data is still being collected and the studies are still being designed. The National
Children’s Study in the U.S., which is being conducted by the U.S. National Institute of Child
Roy & Moser Jones
24


Health and Developm
ent (NICHD) in collaboration with the Environmental Protection Agency
(EPA) and U.S. Centers for Disease Control and Prevention (CDC) is prospectively following a
volunteer cohort of over 100,000 children from 105 geographic areas in the U.S. for 21 years.

It
is collecting data on a variety of environmental exposures for these children, through
a

structured
interview process as well as through collection of biological and environmental samples. The
study aims to “support improved understanding of the ways t
he social and physical environment,
in combination with other exposures, affects children’s development.” Some of the other factors
include “language, culture, change in location, travel, access to health care, participation in
organized activities, recrea
tional opportunities, structured learning, out of home child care, and
community involvement.” (National Children’s Study, 2011) Currently, the study is at the
vanguard
(pilot)
phase. The current study design does not include collection of data on measure
s

of family environment or well
-
being: only data on maternal and paternal education, SES,

and
housing is being collected

(Trasande,
et al.
, 2009). However, the study is still “evolving”
and
the
main phase of the study has not yet begun
.

As a

federally suppo
rted

study
, it is subject to input
from the wider research community,

including scientists who may want to examine data on the
long
-
term relationships between family
-
level factors, environmental exposures, and children’s
health.

In Australia and the United

Kingdom, two similarly relevant large
-
scale prospective
cohort studies, the Environments for Healthy Living
(EHL) st
udies, are being conducted to look
at the relationships between genetics, family, and the environment. The studies
follow

mother
-
child dyad
s followed from
the child’s
birth. The data is being collected in the form of cord blood,
questionnaires self
-
administered by the mothers on “familial, social, and environmental
exposures”, and medical records. The familial exposures include maternal behav
iors such as
Roy & Moser Jones
25


smoking and substance use,
and
maternal mental health indicators as well as family structures,
income, and social support. The short form of the Family Environment scale and the Li
fe Events
Stress Scale are

included with the maternal questionn
aires. For researchers who wish to study
family
-

environment interactions, these datasets may provide
a
valuable resource

(Cameron
et
al.
, 2012).

But they do not track a U.S. cohort.

If the National Children’s study or another
longitudinal study were to add variables related to family processes, possibilities for research on
f
amily
-
environment interactions could be expanded
.


Health interventions,
family literacy
, and disparities

An i
ntegrated life course framework models how social change is more than the cluster of
social determinants of health and well
-
being. But how is it
more
, and what can that mean

for
mobilization to intervene to promote better health in an era of rising inequa
lity?
The framework
offers tools but also challenges to discover new patterns, new pathways,

and new relationships
that may

result in better understanding and more effective interventions to secure healthy
families and communities.

A life course approach
may encourage public health and family science to explore
the
vision of health literacy and

how family members can promote or discourage literacy in a
household or community
.
It would also issue practical challenges of how to translate theory
into pract
ice. Perhaps a life course framework is not theoretical enough
. But this supposed
disadvantage would could be turned on its head to
offer both fields a more flexible set of
assumptions and methods that can lend themselves more readily to application.

P
ublic health and family science can both utilize a life course framework to
implement
critical interventions

to promote greater equity in health and well
-
being for individuals and their
Roy & Moser Jones
26


families.
This goal has been a long
-
standing, driving force for mobil
ization of resources in
public health, particularly in maternal and child health. In contrast, family science lacks such a
compelling agenda for change and clear resources to provide “deliverables” to families in
communities. Partnership with public heal
th, and through a life course paradigm, may nudge
family scientists to take clear positions on issues of health disparities and related challenges to
social equity.
The life course lens, specifically, may draw attention to the consequences of
childhood d
isease and poverty, the experience of accumulated disadvantage in old age, and the
differential health impacts of economic recession on families of color, on women, and on poor
families.


Conclusion

Whe
n and where

does health become a family issue?

Using

the life course framework to
explore this question
would open up consideration of what happens between populations and
individuals. It would challenge core assumptions that health issues are focused on children and
women, and not men


and on individuals
, and not networks of family and friends. It would
push researchers and practitioners to develop more finely
-
tuned links between biology and
society (Shanahan & Boardman, 2009), which is a massive project stretching across the social
and physical sciences
.

A life course framework, we argue, may be the first of many bridges constructed between
public health and family science. We have indicated a number of potential points for futu
re
collaboration and growth. Over the course of

this project, we develope
d a metaphor that fit our
discovery of what life course could do for both of these fields. Public health can be considered
as a large institutional building, well
-
established and packed with many floors of rooms, each
Roy & Moser Jones
27


populated by researchers, practitione
rs, community actors, physicians, and social workers. There
is a series of rooms that are empty, however, and await experts in family life. If family scientists
enter into this building, they will take advantage of

its

institutional
capacity

to mobilize
its vast
resources
around family health concerns.

Similarly, family science can be considered as an array of individual family households
scattered throughout a community. Each household has unique dynamics and family
relationships, but many of the same i
ssues and processes emerge across this array: partnering
and parenting, work and family stress, children growing into adolescence, multiple generations
living under the same roof. Public health researchers need to leave their institution and travel to
th
ese households, to witness these processes and how they might shape health outcomes for
women, children, and men who live there.


And the emphasis on time? First, it will take time for these collaborations to unfold in
research and practice. But more imp
ortantly, researchers in both fields recognize that a clear
understanding of health in families remains obscured without repeated visits, over time, to
capture how
the processes of
growth and decline actually

occur in the context of daily life
.



Roy & Moser Jones
28


Referen
ces


Baltimore Healthy Start website.
Our Mission
. Retrieved October 18, 2012, from
http://www.baltimorehealthystart.org/
.

Bauldry, S., Shanahan, M., Boardman, J., Miech, R., &

Macmillan, R. (2012). A life course model of

self
-
rated health through adolescence and young adulthood.
Social Science Medicine, 75
, 1311
-
1320.

Beaglehole, R., and Magnus, P. (2002). The search for new risk factors for coronary heart disease:
occupa
tional therapy for epidemiologists?
International Journal of Epidemiology
, 31, 6, 1117
-
1122.

Ben
-
Shlomo, Y., & Kuh, D. (2002). A Life Course Approach to Chronic Disease Epidemiology:
Conceptual Models, Empirical Challenges and Interdisciplinary Perspective
s.
International
Journal of Epidemiology
, 31, 285
-
293.

Ben
-
Shlomo,

Y
. (2007). Rising to the Challenges and Opportunities of Life Course Ep
idemiology.
International Journal of Epidemiology
, 36, 481
-
483.

Bengston, V. & Allen, K. (1993). The life course perspective applied to families over time. In


P. Boss, W. Doherty, R. LaRossa, W. Schumm, & S. Steinmetz (Eds.),
Sourcebook of family


theories and methods: A contextual approach
(pp. 469
-
499). New York: Plenum.

Bengtson, V., Rosenthal, C., & Burton, L. (1990). Families and aging: Diversity and heterogeneity. In ]

R. Binstock & L. George (eds.),
Handbook on aging and the social
sciences

(3
rd

ed., pp. 263
-

287). San Diego: Academic Press.

Boehmer, U., Miao, X., Linkletter, C., & Clark, M. (2012). Adult health behaviors over the life

course by sexual orientation.
American Journal of Public Health, 102
, 292
-
300.

Burton, L., Pu
rvin, D., & Garrett
-
Peters, R. (2009). Longitudinal ethnography: Uncovering

domestic abuse in low
-
income women’s lives. In G. Elder, Jr., & J. Giele (eds.),
The craft of life
course research
. New York: Guilford Press.

Cameron, C., Scuffham, P., Spin
ks, A., Scott, R., Sipe, N., Ng, S., Wilson, A., Searle, J., Lyons, R.,
Kendall, E., Halford, K., Griffiths, L., Homel, R., & McClure, R. (2012). Environments for
Healthy Living (EFHL) Griffith Birth Cohort Study: Background and Methods.
Maternal and
Child

Health Journal
, 7 Feb (epub).

Carr, D. & Springer, K. (2010). Advances in families and health research in the 21
st

century.
Journal of

Marriage and Family,
72
, 743
-
761.

D’Onofrio, B., &

Lahey, B. (2010). Biosocial influences on the family: A decade review.
Journal of

Marriage and Family, 72
, 762
-
782.

Davey Smith, G. (2007).
Life
-
course

Approaches to Inequalities in Adult Chronic Disease Risk.
The
Proceedings of the Nutrition Society
, 66, 216
-
36.

Deardorff,

J
,
Berry
-
Millett, R
,
Rehkopf, D
.,
Luecke ,E
.,
Lahiff, M
.,
Abrams, B
. (2012). Maternal Pre
-
pregnancy BMI, Gestational Weight Gain, and Age at Menarche in Daughters.
Maternal and
Child Health Journal
, Oct 2 (Epub).

Denney, J. (2010). Family and household formation and suicide in the United States.
Journal of


Marriage and Family, 72
, 212
-
213.

Doll R., Hill A.B., (1954). The mortality of doctors in relation to their smoking habits.
British Medical
Journal,
328
, 1529.

Doll, R.; Hill A.B. (1956). Lung cancer and other causes of death in relation to smoking; a second report
on the mortality of British doctors.
British Medical Journal

2
, 5001, 1071

1081.

Doll, R.; Hill A.B. (1964).
Mortality in Relation to Smoking: Ten Ye
ars' Observations of British Doctors.
British Medical Journal, 1, 5396,
1460

1467.

Eckenrode, J., Campa, M., Luckey, D., Henderson, C., Cole, R.; Kitzman, H., Anson, E., Sidora
-
Arcoleo,
K., Powers, J., & Olds, D. (2010).
Long
-
term Effects of Prenatal and
Infancy Nurse Home
Visitation on the Life Course of Youths:
19
-
Year Follow
-
up of a Randomized Trial
.

Archives of
Pediatric and Adolescent Medicine
,
164
, 9
-
15.

Roy & Moser Jones
29


Elder, G.H., Jr. (1996). Human lives in changing societies: Life course and developmental ins
ights. In

R. Cairns, G.H. Elder Jr., & E. Costello (Eds.),
Developmental science

(pp. 31
-
62 ). New York:
Cambridge University Press.

Elder, G.H., Jr. (1999
).

Children of the Great Depression: Social change in life experience


(25th anniversary ed.).

Boulder, CO: Westview Press.

Elder, G., Jr., & Taylor, M. (2009). Linking research questions to data archives. In G. Elder,

Jr., & J. Giele (eds.),
The craft of life course research
. New York: Guilford Press.

Epstein, M., Hill, K., Bailey, J., & Haw
kins, J. (2012). The effects of general and drug
-
specific family

environments on comorbid and drug
-
specific problem behavior: A longitudinal examination.
Developmental Psychology
, July.

Fine, A., & Kotelchuck, M. (2010).
Rethinking MCH: The life course model as an organizing

framework.
Washington DC: US Department of Health and Human Services, Health Resources
and Services Administration, Maternal and Child Health Bureau.

Fine, M. Kotelchuck, M., Adess, N., Pies, C.

(2009). Policy Brief A New Agenda for MCH Policy and
Programs: Integrating A Life Course Perspective.

Family, Maternal and Child Health Programs.
Contra Costa, CA,
Contra Costa Health Services,

Fleten,

C
.,
Nystad, W
.,
Stigum, H
.,
Skjaerven, R
.,
Lawlor, DA
.,
Davey Smith, G
. &
Naess ,O
. Parent
-
offspring body mass index associations in the Norwegian Mother and Child Cohort Study: a
family
-
based approach to studying the ro
le of the intrauterine environment in childhood
adiposity.
American Journal of Epidemiology
, 176, 83
-
92.

Gamborg, M., Jensen, G., Sorensen, T., & Andersen, P. (2011). Dynamic Path Analysis in Life
-
Course
Epidemiology.
American Journal of Epidemiology
,
173
, 10, 1131
-
1139.

George, L.K. (2009). Conceptualizing and measuring trajectories. In G. Elder, Jr., & J. Giele

(eds.),
The craft of life course research
. New York: Guilford Press.

Geronimus, A., Hicken, M., Keene, D., &

Bound, J. (2006). Weathering
and age patterns of allostatic

load scores among Blacks and Whites in the United States.
American Journal of Public Health,

96
, 826
-
833.

Giele, J., & Elder, G.H., Jr. (1998).
Methods of life course research: Qualitativ
e and

quantitative approaches.

Thousand Oaks, CA:
Sag
e.

Giles, L., Davies, M., Whitrow, M., Rumbold, A., Lynch, J., Sawyer, M., Moore, V. (2011). Structured
Regression Analyses of Life Course Processes: An Example Exploring How Maternal Depression
in Ea
rly Childhood Affects Children’s Subsequent Internalizing Behavior.
Annals of
Epidemiology
, 21:654

659.

Guo, G., Roettger, M., & Cai, T. (2008). The integration of genetic propensities into social
-
control

models of delinquency and violence among male yo
uths.
American Sociological Review, 73
,

543
-
568.

Hallqvist, J
.,
Lynch, J
.,
Bartley, M
.,
Lang, T
.,
Blane, D
. (2004). Can we disent
angle
life course

processes
of accumulation, critical period and social mobility? An analysis of disadvantaged socio
-
economic positions and myocardial infarction in the Stockholm Heart Epidemiology Program.
Social Science & Medicine
, 58, 1555
-
62.

Hauser, R
. (2009). The Wisconsin Longitudinal Study: Designing a study of the life course. In

G. Elder, Jr., & J. Giele (eds.),
The craft of life course research
. New York: Guilford Press.

Halfon, J., & Hochstein, M. (2002). Life course health development:

An integrated framework

for developing health, policy, and research.
Milbank Quarterly, 80
, 433
-
479.

Hess, R., & Handel, G. (1959).
Family worlds.

Chicago: University of Chicago Press.

Higgins, C., Duxbury, L., &

Lyons, S. (2010). Coping with overload and stress: Men and women in

dual
-
earner families.
Journal of Marriage and Family, 72
, 847
-
859.

Institute of Medicine, (1988).
The Future of Public Health.

Washington, D.C., National Academies

Press.

Roy & Moser Jones
30


J
okela,

M
.,
Power C
.,
Kivimäki, M
. (2009) Childhood problem behaviors and injury risk over the
life

course
.
Journal of Child Psychology and Psychiatry
, 50, 1541
-
9.

Jones, M.M. (2011). Race, Class and Gender Disparities in Clara Barton’s Late Nineteenth
-
Century
Disaster Relief.
Environment and History
,
17,

107
-
131.

Jones, M.M. (f2013).
The American Red Cross
-

From Clara Barton to the New Deal.
Baltimore, MD:
Johns H
opkins University Press.

Juon, H., Fothergill, K., Green, K., Doherty, E., & Ensminger, M.
(2012).
Antecedents and

consequences of marijuana use trajectories over the life course of an African American
population. Department of Health, Behavior and Soc
iety, Johns Hopkins Bl
oomberg School of
Public Health.

Kelsey
, J.
,
Whittemore
, A., Evans, A., & Thompson, W. (1996).
Methods in Observational Epidemiology
.
New York,
Oxford University Press.

Kerr, D., Capaldi, D., Owen, L., Wiesner, M., & Pears, K. (2011). Changes in at
-
risk American men’s

crime and substance use trajectories following fatherhood.
Journal of Marriage and Family, 73
,
1101
-
1116.

Kim J
., &
Richardson

V
. (2012). The impact of socioeconomic inequalities and lack of health insurance
on physical functioning among middle
-
aged and older adults in the United States.
Health and

Social Care in the Community
, 20, 42
-
51.

Kuh DL, Ben
-
Shlomo Y. (1997).
A Life Cours
e Approach to Chronic Disease Epidemiology; Tracing the
Origins of Ill
-
health from Early to Adult Life
. Oxford, Oxford University Press.

Kuh, D., Ben
-
Shlomo,
Y., Lynch, J., Hallqvist, J., & Power, C. (2003). Glossary: Life Course
Epidemiology.
Journal of E
pidemiology and Community Health, 57
, 778

783.

Laub, J., & Sampson, R.

(2003).
Shared beginnings, divergent lives: Delinquent boys to age 70
.


Boston: Harvard University Press.

Li., Q., &

Keith, L. (2011). The differential association between education and infant mortality

by nativity status of Chinese American mothers: A life course perspective.
American Journal of
Public Health, 101
, 899
-
908.

Lu M. & Halfon, N. (2003). Racial and
ethnic disparities in birth outcomes: a Life
-
course Perspective.
Maternal and Child Health Journal, 7, 13

30.

Lu, M.,
Kotelchuck, M
.,
Hogan, V
.,
Jones, L
.,
Wright, K
.,
Halfon, N
. (2010). Closing the Bl
ack
-
White
gap in Birth Outcomes: a
Life
-
Course

Approach.
Ethnicity & Disease
, 20(1 Suppl 2), S2
-
62
-
76.

Lynch

J
., Kaplan, G.,
Salonen, R
.,
Salonen, JT
. Socio
economic status and progression of carotid

atherosclerosis. Prospective evidence from the Kuopio Ischemic Heart Disease Risk Factor Study
.
Arteriosclerosis, Thrombosis & Vascular Biology
, 17, 513
-
519.

Lynch, J. & Davey Smith, G. (2005)
A Life Course Approach to Chronic Disease Epidemiology.
Annual
Reviews in Public Health
,
26
, 1

35.

Maring, E., Singer, B., & Shenassa, E. (2010). Making the transition from lead based prevention to

healthy homes: A qualitative study
. Journal of Publi
c Health Management and Practice, 16
,
S53
-
S60.

Marsiglio, W., & Roy, K. (
2012
) Fathers’ nurturance of children over the life course. In G. Petersen &

K. Bush (eds.),
Handbook of Marriage and the Family
(3
rd

ed)

(pp. 353
-
376)
. New York:
Springer.

Mi
ller, W., Sadegh
-
Nobari, T., & Lillie
-
Blanton, M. (2011). Healthy Starts for All: Policy Prescriptions.
American Journal of Preventive Medicine
, 40(1S1), S19

S37.


Moen, P., &

Chermack, K. (2005). Gender disparities in health: Stategic selection, careers, and cycles

of control
. Journal of Gerontology, 60
, S99
-
S108.

Montoro
-
Rodriguez, J., Smith, G., & Palmieri, P. Use of community and school mental health services

by custodial grandchildren.
Family Relations, 61
, 207
-
223.

Moser Jones, M. (forthcoming).
The American Red Cross
-

From Clara Barton to the New Deal.

Baltimore, MD:
Johns Hopkins University Press.

Roy & Moser Jones
31


Najman, J., Hayatbakhsh, M., Clavarino, A., Bor,

W., O’Callaghan, M.,. Williams, G. (2010). Family
Poverty over the Early Life Course and Recurrent Adolescent and Young Adult Anxiety and
Depression: A Longitudinal Study.
American Journal of Public Health

100,
1719
-
1723.

O’Rand, A. (2009).
Cumulative p
rocesses in the life course. In G. Elder, Jr., & J. Giele (eds.),

The craft of life course research
. New York: Guilford Press.

Owen

CG
.,
Martin RM
.,
Whincup, PH
.,
Smith GD
.,
Cook, DG

(2005). Effect

of
infant

feeding

on the risk
of obesity across the
life

course
: a quantitative review of published evidence. Pediatrics, 115,
1367
-
77.

Pies, C., Parthasarathy, P., & Posner, S. (2012). Integrating the life course perspective into a local

maternal and
child health program.
Maternal and Child Health Journal, 16
, 649
-
655.

Richardson, J. (2011). At wit’s end: A qualitative study of the relationship between African American

parenting strategies, social capital, the juvenile justice system, and mental

health care for at
-
risk
youth.
Journal of Correctional Healthcare, 17
, 319
-
328.

Roustit, C., Campoy, E., Renahy, E., King, G., Parizot, I., Chauvin, P. Family Social Environment in
Childhood and Self
-
Rated Health in Young Adulthood. (2011).
BMC Public Health, 11,
949.

Roy, K. (2006). Father stories: A life course examination of paternal identity among


low
-
income African American men.
Journal of Family Issues, 27
, 31
-
54.

Roy, K. (2008). A life course perspective on fatherhood and fami
ly policies in the United States

and South Africa.
Fathering: A Journal of Theory, Research, and Practice about Men as
Fathers, 6,
92
-
112.

Schmeer, K. (2011). The child health disadvantage of parental cohabitation.
Journal of Marriage and

Family, 73
,
181
-
193.

Shanahan, M., & Boardman, J. (2009). Genetics and behavior in the life course: A promising

frontier. In G. Elder, Jr., & J. Giele (eds.),
The craft of life course research
. New York:
Guilford Press.

Slep, A, Foran, H., Heyman, R., & Sna
rr, J. Risk factors for clinically significant intimate partner

violence among active
-
duty members
. Journal of Marriage and Family, 73
, 486
-
501.

Smith, K., & Christakis, N. (2008). Social networks and health.
Annual Review of Sociology, 34
, 405
-

429.

Steingraber, S. (2011).
Raising Elijah: Protecting our children in an age of environmental crisis
.

Cambridge, MA: De Capo Press.

Trasande, L., Cronk, C., Durkin, M., Weiss, M., Schoeller, D., Gall, E., Hewitt, J., Carrel, A., Landrigan,
P., & G
illman, M. (2009). Environment and Obesity in the National Children’s Study.
Environmental Health Perspectives
,
117
, 159
-
166.

Turney, K., & Carlson, M. (2011). Multipartnered fertility and depression among fragile families.

Journal of Marriage and Fami
ly, 73
, 570
-
587.

Wadhwa, PD
,
Bu
ss, C
,
Entringer, S
,
Swan
son, J.M
. (2009). Developmental origins of health and disease:

brief history of the approach and current focus on epigenetic mechanisms.

Seminars in
Reproductive Medicine
, 27, 358
-
68.

White, J. (2005
).
Advancing family theories
. Thousand Oaks, CA:
Sage.

White, J., & Klein, D. (2008).
Family theories

(3
rd

ed). Thousand Oaks, CA: Sage.

Wickrama, K., Surjadi, F., Lorenz, F., Conger, R., & O’Neal, C. (2012). Family economic hardship and

progression of poor mental health in middle
-
aged husban
ds and wives.
Family Relations, 61
,

297
-
312.

Wiersma, J., Harrington, C., Herrera, V., & Fischer, J. (2010). Intimate partner violence in young adult

dating, cohabiting, and married drinking partnerships
. Journal of Marriage and Family, 72
, 360
-
374.

Yu, S. (2006). The life course approach to public health.
American Journal of Public Health, 96,

768.

Zhang, Z., & Hayward, M. (2006). Gender, the marital life course, and cardiovascular disease in late

midlife.
Journal of Marriage and Family, 68
,
639
-
657.