Accepted Manuscript

Childhood Adversity and Adult Health: Evaluating Intervening Mechanisms

R. Jay Turner, PhD, Courtney S. Thomas, PhD, Assistant Professor of Sociology/African American & Africana Studies, Tyson H. Brown, PhD

PII: S0277-9536(16)30081-8

DOI: 10.1016/j.socscimed.2016.02.026

Reference: SSM 10527

To appear in: Social Science & Medicine

Received Date: 9 April 2015

Revised Date: 14 February 2016

Accepted Date: 16 February 2016

Please cite this article as: Turner, R.J., Thomas, C.S., Brown, T.H., Childhood Adversity and AdultHealth: Evaluating Intervening Mechanisms, Social Science & Medicine (2016), doi: 10.1016/j.socscimed.2016.02.026.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Childhood Adversity and Adult Health: Evaluating Intervening Mechanisms*

R. Jay Turner, PhD Vanderbilt University

Courtney S. Thomas, PhD1

University of Kentucky

Tyson H. Brown, PhD Vanderbilt University

1Corresponding Author: Courtney S. Thomas, PhD, Assistant Professor of Sociology/African American & Africana Studies, University of Kentucky, 1515 Patterson Office Tower, Lexington, Kentucky 40506-0027. Phone: 859-257-4414. Email: courtney.s.thomas@uky.edu *This research was supported by a grant (R01AG034067) from the Office of Behavioral and Social Science Research and the National Institute on Aging to R. Jay Turner.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Childhood Adversity and Adult Health: Evaluating Intervening Mechanisms

ABSTRACT Substantial evidence has accumulated supporting a causal link between childhood adversity and risk for poor health years and even decades later. One interpretation of this evidence is that this linkage arises largely or exclusively from a process of biological embedding that is not modifiable by subsequent social context or experience—implying childhood as perhaps the only point at which intervention efforts are likely to be effective. This paper considers the extent to which this long-term association arises from intervening differences in social context and/or environmental experiences—a finding that would suggest that post-childhood prevention efforts may also be effective. Based on the argument that the selected research definition of adult health status may have implications for the early adversity-adult health linkage, we use a representative community sample of black and white adults (N=1252) to evaluate this relationship across three health indices: doctor diagnosed illnesses, self-rated health, and allostatic load. Results generally indicate that observed relationships between estimates of childhood adversity and dimensions of adult health status were totally or almost totally accounted for by variations in adult socioeconomic position (SEP) and adult stress exposure. One exception is the childhood SEP-allostatic load association, for which a statistically significant relationship remained in the context of adult stress and SEP. This lone finding supports a conclusion that the impact of childhood adversity is not always redeemable by subsequent experience. However, in general, analyses suggest the likely utility of interventions beyond childhood aimed at reducing exposure to social stress and improving social and economic standing. Whatever the effects on adult health that derive from biological embedding, they appear to be primarily indirect effects through adult social context and exposure. Keywords: USA; childhood adversity; allostatic load; self-rated health; chronic illness; life course; stress; trauma

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

1

Childhood Adversity and Adult Health: Evaluating Intervening Mechanisms

In a relatively recent publication of immense potential significance, a highly distinguished

group of researchers have argued that there is now a substantial and growing body of evidence

indicating that adult health disparities have their roots in childhood adversity (Shonkoff, Boyce &

McEwen, 2009). This body of research has been most effectively reviewed by Miller, Chen, &

Parker (2011) who judge the association to be robust and likely causal in nature. Shonkoff et al.,

(2011) see this evidence as strongly suggesting that reducing exposure to early life adversity may

be required to effectively address adult health disparities and, thus, should be the focus of

prevention and intervention efforts—a policy implication that may or may not be fully justified.

The question of the extent to which adult health disparities are largely and immutably

fashioned in childhood or redeemable by subsequent social context and experience is of

substantial importance because the answer is crucial for understanding the origins of adult health

inequalities and for estimating the likely utility of post-childhood intervention efforts. This paper

attempts to contribute toward such an answer with a recently completed population-based study

that offers a unique opportunity to more fully consider this important question. Available data

allows for comprehensive assessment of early and adulthood adversity, adult socioeconomic

position (SEP), and multiple indicators of adult health that include both self-report and biomarker

estimates. With these advantages, we consider the extent to which the effects of early adversity

on adult health work primarily through subsequent stress exposures and/or social structural

experience. As we will show, childhood experiences are largely, though not totally, mediated by

SEP and exposure to social stress in adulthood.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

2

BACKGROUND

As Miller et al. (2011) have noted, almost all studies linking early stress to adult health

have focused on either parental maltreatment or childhood socioeconomic disadvantage. That

these two streams of evidence appear to converge adds confidence that there is a meaningful

linkage between childhood adversity and adult health. However, the meaning of these reliable

linkages remains to be established. As suggested above, one prominent hypothesis is that this

association arises largely or exclusively from an embedding process of some kind, with long-term

health implications that are not modifiable by subsequent social context and experience (Miller et

al., 2011). If confirmed, such a conclusion would indeed suggest that early adversity has fateful

health implications and that childhood may be the only point at which intervention efforts are

likely to be effective in reducing adult health disparities. This version of embedding is thought to

occur at the molecular level and involve enduring processes such as those described by Miller et

al. (2011). A second version of this hypothesis argues that the embedding or scarring resulting

from childhood adversity sets limits on the individual’s capacity to develop effective coping

strategies and/or to gain and maintain supportive social relationships. That early adversity can

have such long-term consequences for social support has recently been demonstrated by

Umberson and colleagues (2014).

A sharply contrasting point of view is that the early adversity-adult health relationship is

best understood from a chains of risk perspective that sees early adversity as a major risk factor

for future adversity, with greater accumulation resulting in elevated health risk (Blane, 1999; Kuh

& Ben-Shlomo, 2004; Lynch & Smith, 2005; Pollitt, Rose & Kaufman, 2005). A crucial corollary

of this perspective is that this significant over-time association arises largely from potentially

modifiable intervening differences in social context and/or environmental experience and,

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

3

accordingly, intervention efforts over the life course are both justified and likely to be effective.

The chains of risk hypothesis differs markedly from the second embedding hypothesis in that

adult variations in health-relevant contingencies are attributed to biological rather than social

origins. Because the factors hypothesized to intervene between early adversity and adult health

overlap across these two perspectives, present data will not allow meaningful adjudication of their

relative plausibility. As such, this paper presents data that examine the relative plausibility of the

first of the embedding hypotheses and the chains of risk perspective.

Although evidence of a meaningful relationship between early adversity and adult health

appears solid, there are several unsettled questions that deserve additional consideration. Among

these is whether measurement of health status within prior research may have yielded either an

under- or overestimate of the health significance of early adversity. In a review of extant animal

and human studies, Cassel (1974, 1976) argued long ago that the social environment acts to raise

or lower risk for all forms of disorder and that the nature of the particular disorders that eventuate

is determined on other grounds. Nevertheless, most studies on the early adversity-adult health

relationship have considered specific individual disorders or narrow ranges of related disorders.

Consequently, such research may involve misclassification error to the extent that 1) early

adversity has effects on health problems and disorders in addition to, or other than, those captured

by, or reflected within, the particular research definition employed, and 2) those classified as

“well” have health problems that have not yet been labeled by a physician or have not yet been

clearly experienced symptomatically.

As Aneshensel (2005; Aneshensel, Rutter, & Lachenbruch, 1991) has suggested, the

misclassification of individuals with unmeasured or undetected forms of illness as not disordered

is likely to have resulted in an under- or over-estimation of the health relevance of the

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

4

environmental factors under study. The crucial point is that early trauma and/or significant

childhood disadvantage may not be linked to a specific disorder, or set of related disorders, to the

exclusion of others. Thus, research that considers the implications of alternative operational

definitions of adult health status may provide a better test of the early adversity-adult health

linkage (Turner, 2010; 2013).

Estimating Adult Health Status

Studies conducted on representative community samples have often faced limited options

for indexing variations in general health status, with many efforts relying on participant reports of

perceived health or doctor-diagnosed health problems. In addition, there is a rapidly growing

literature employing biomarker data to estimate current health status, within which allostatic load

(AL) has received substantial attention. To effectively evaluate the early adversity-adult health

relationship, we utilize three relatively broad and often-employed indexes of physical health

status, the strengths and weaknesses of which are considered below.

Self-Rated Health. The most popular of these operational definitions appears to be self-

rated health. This popularity presumably arises from the convenience and low cost to obtain these

ratings, in addition to evidence of their predictive validity with respect to mortality (Benyamini &

Idler, 1999; Idler, 1993; Idler & Angel 1990), morbidity (Farmer & Ferraro, 1995; Ferraro,

Farmer, & Wybraniec, 1997), and physical disability (Ferraro et al., 1997; Idler & Kasl, 1995).

Moreover, prior studies suggest that these predictive associations do not vary appreciably across

racial groups (Gibson, 1991; Johnson & Wolinsky, 1994).

Although available evidence leaves room for uncertainty, it may be that most instances of

self-reported doctor-diagnosed disorders, diseases, and physical limitations tend to be reflected in

reports of fair or poor health (Hardy, Acciai, & Reyes, 2014). However, it may also misclassify

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

5

individuals as ill who are not, thereby diluting the power of the disordered-non-disordered

contrast. An additional concern is the possibility of state dependence, with individuals

experiencing emotional distress at the time of the rating being more likely to rate their physical

health as fair or poor.

Self-Reported Doctor Diagnosed Illness. Many studies have distinguished the ill from the

well by questioning respondents on whether a doctor had told them they had life- threatening

problems such as heart disease, stroke, cancer, hypertension, COPD, diabetes and/or one or more

of other specific chronic diseases. Disorders reported are considered individually (e.g. Hayward,

Crimmins, Miles, & Yang, 2000) or as counts of reported disorders (e.g. Sternthal, Slopen &

Williams, 2011; House, Lepkowski, Kinney et al., 1994; Ross & Wu, 1995). Underlying this

approach are two necessary assumptions: 1) that people can reliably report information provided

by their doctor and 2) that occurrences of virtually all diseases and disorders reliably come to the

attention of doctors. However, it is clear that there are racial and socioeconomic differences in the

availability and/or the utilization medical care (e.g. Escarce et al., 1993; Ferguson et al., 1997;

Fincher et al., 2004; Peterson et al., 1997; Klabunde et al., 1998) and in level of trust in health

care institutions and physicians (e.g. Kao et al., 1998a, 1998b; Saha et al., 2003). Consequently,

African Americans, men, and those of low SEP who are ill may be disproportionately counted as

well. This form of bias also characterizes studies employing objective measures of adult health

through use of hospital or clinic records. While it seems safe to conclude that research definitions

of health in this category are biased, the extent of this bias and its implications for estimating the

magnitude of the early adversity-adult health relationship are unknown.

Allostatic Load (AL). McEwen and colleagues (McEwen & Steller, 1993; McEwen, 1999;

McEwen & Seeman, 1999) formulated the concept of AL referring to the cumulative wear and

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

6

tear on the body‘s systems owing to repeated adaptation to stressors (Geronimus, Hicken, Keene,

& Bound, 2006). Allostatic load is thus thought to provide a meaningful description of the long-

term biological consequences of chronic stress (McEwen and Seeman, 1999). Such stress

exposure results in dysregulation that is reflected by a change in the set- points of physiological

markers (Dowd and Goldman 2006). Geronimus et al. (2006:826) highlight advantages of the

allostatic load measure, including its assessment of multiple physiological systems, including

responses that increase the risk of morbidity and mortality, but may have not yet registered

clinically (Karlamangla et al., 2006; Seeman et al., 2004). Adversity during childhood may trigger

or imbed biological processes that result in dysregulation. Alternatively, early adversity may

sensitize the individual to later exposures to social stress or may simply foreshadow high levels of

later exposure.

As a result, there appear reasonable grounds for the assumption that allostatic load can be

taken as a meaningful estimate of health status at the point of biomarker collection. The particular

strengths of this measure as an operational definitions of adult health status include: 1) it does not

rely on equality across statuses in health service availability, utilization, or adequacy of doctor-

patient communication, 2) it depends on neither memory nor perception, and 3) it is consistent

with the increasingly accepted view that the social environment acts to raise or lower risk for all

kinds of disease and disorder in general. The examination of these three indices of adult physical

health status represent an important step toward reducing misclassification bias, which may allow

for the more adequate consideration of the early adversity-adult health association.

Estimating Level of Adversity

A second question worthy of consideration is whether childhood major and potentially

traumatic events beyond, or other than, parental maltreatment are consequential for adult health.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

7

As noted above, Miller et al. (2011) acknowledge that although there are many severe stressors to

which children might be exposed, nearly all of the extant research linking early stress to adult

health has focused on one of two experiences: parental maltreatment or socioeconomic

disadvantage. While much of the evidence for this association has been reported in this century,

the long-term consequences and health relevance of parental maltreatment is hardly so new.

There is a substantial body of evidence that a bad start in terms of life conditions and the

quality of parenting is predictive of poor physical, emotional, and social development, all of

which are relevant for adult health (Repetti, Taylor, & Seeman, 2002). Parallel with this relatively

long-standing literature are findings of linkages between psychiatric and substance disorders in

adulthood and childhood abuse (Browne & Finkelhor, 1986; Burnam, et al., 1988; Bryer, Nelson,

Miller, & Krol, 1987; Wisdom et al., 1999; Kessler & Magee, 1993). Importantly, similar results

have been reported in connection with parental death and parental mental health and substance

use problems, suggesting the significance of traumatic experiences other than parental

maladaptation (Brown, Harris, & Bifulco, 1986; Mcleod, 1991; Weissman, Gammon, Merikangas

et al., 1987). Consistent with these findings is evidence that there is a range of major and

potentially traumatic stressors that, singly or in combination, have been shown to predict the onset

of drug dependence, alcohol dependence, and major psychiatric disorders, in addition to increased

risk that a subsequent stressor will eventuate in PTSD (Lloyd & Turner, 2003, 2008; Turner &

Lloyd, 1995, 2003, 2004). Thus, some occurrences can and do have significant mental health

consequences despite occurring years or even decades earlier, suggesting the possibility of

parallel long-term physical health consequences.

In addition to early exposure to social stressors, childhood socioeconomic position (SEP)

has also been shown to predict adult health and all-cause mortality—relationships that have

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

8

generally persisted when adult SEP was controlled (Galobardes, Lynch, & Smith, 2004, 2006,

2008). However, the comprehensive measurement of both childhood and adult SEP is relatively

uncommon in prior research. For example, a number of studies reporting the association have

estimated childhood SEP using only father‘s occupational level and adult SEP only in terms of the

individuals own occupational level (Miller et al., 2011). More recently, several studies find that

low levels of parental education and financial hardship have deleterious effects on adult health

(Brown, O‘Rand, & Adkins, 2012; Haas, 2008; Haas & Rohlfsen, 2010; Warner & Brown, 2011).

Moreover, fully comprehensive measures of childhood SEP have been employed in newer studies

predicting mortality (Hayward & Gorman, 2004; Warner & Hayward, 2006) and disability (Haas,

2007; Haas & Rohlfsen, 2009; Warner & Brown, 2011). Consequently, relevant aspects of

childhood context and adult disadvantage may not have been fully captured within earlier

research establishing an association between early adversity and adult morbidity.

From their wide-ranging review of evidence, Miller et al. (2011: 962) also underscore the

need for more attention to what transpires in the period between exposure to early adversity and

later manifestations of disease. Based on research accumulated over the past 25 years indicating

that both mental and physical health risk are substantially conditioned by the social contexts and

the stress experiences to which the individual has been and is being exposed, there are solid

grounds for viewing the relationship at issue as reflecting the tendency for high levels of

childhood stress to predict elevated levels of later exposure as well as health status in adulthood.

Thus, parental maltreatment, other traumatic exposure, and low childhood SEP may simply be

effective markers of lives subsequently filled with stressors. However, evidence effectively

assessing the possibility that it is lifetime exposure to social stress rather than stress in childhood

that matters for adult health risk appears largely absent from the literature. This is so despite

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

9

long-standing evidence linking exposure to social stress and both the onset and persistence of

numerous potentially fatal and chronic health problems including cardiovascular disease (Jenkins

et al., 1978; Kaplan et al., 1982; Kaplan & Keil, 1983; Nerem, Levesque, & Cornhill, 1980;

Rozanski, Blumenthal, &Kaplan, 1999; Vitaliano et al., 2002); multiple sclerosis (Grant et al.,

1989; Stip & Truelle, 1994; Warren, Greenhill, & Warren, 1982); diabetes mellitus (Hagglof et

al., 1991; Leaverton et al., 1980; Mooy et al., 2000; Thernlund et al., 1995); high blood pressure

(Karlsen & Nazroo, 2002; Krieger & Sydney, 1996); fibromyalgia (Kivimaki et al., 2004);

rheumatoid arthritis and osteoarthritis (Rogers et al., 1980; Zautra et al., 1994); Graves‘ thyroid

disease (Harris et al.,1992; Kung, 1995; Sonino et al., 1993; Winsa et al., 1991); respiratory

illness (Cohen et al., 1999; Cohen et al., 2002; Karlsen & Nazroo, 2002); and both tuberculosis

and cancer (Holmes & Matsuda, 1975).

Within this literature, strategies for assessing differences in stress exposure have varied in

both form and comprehensiveness. However, a clear majority of these studies have indexed such

exposure in terms of recent life events. That this strategy may be inadequate is suggested by

evidence that it underestimates the mental health significance of social stress, as well as the

elevation in stress exposure among African Americans relative to whites and among persons of

lower SEP relative to their more advantaged counterparts (Taylor & Turner, 2002; Turner &

Avison, 2003). The question of whether and the extent to which variations in social stress during

adulthood explains the apparent linkage between childhood adversity and adult health remains to

be adequately addressed and an effective test of this competing hypothesis requires expanded

consideration of the sources or types of stressors to which individuals have been exposed.

From a “chains of risk” perspective, we hypothesize that childhood trauma and/or

childhood SEP are linked to adult health because they are predictive of subsequent stress exposure

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

10

and/or of level of socioeconomic achievement. In this study, we examine the relationships

between adult health and both childhood SEP and early traumatic exposure to determine whether

these associations are wholly or largely artifacts of their joint associations with adult SEP and/or

intervening levels of stressor exposure. In comparison to previous research on adult morbidity,

our efforts to further confirm the early adversity-adult health connection and rule out this

competing hypothesis involve more comprehensive estimates of childhood SEP, adult SEP, and

level of post-childhood stress exposure.

This paper revisits the relationship between early adversity and adult health armed with

several advantages that may collectively provide an advance over prior research. These include:

1) addressing the misclassification problem through the consideration of three distinct research

definitions of current health status, 2) estimating childhood adversity in terms of both low

socioeconomic position and exposure to trauma beyond parental maladaptation, 3) utilization of

multidimensional measures of childhood and adult SEP, and 4) the rather unique consideration of

the intervening effects of multiple dimensions of adult stress exposure. Our primary goal is to

provide new information on whether the childhood adversity-adult health risk linkage reflects a

process of biological embedding that can only be effectively addressed through intervention

during childhood, or represents health-relevant continuities of social contexts and experiences that

are potentially modifiable throughout the life course.

METHOD

Sample

This study was conducted within a midsized southern county. To obtain the study sample,

we used simple random sampling to draw 199 block groups from within the county. Survey

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

11

Sampling International Corporation then provided us with 7,000 randomly selected addresses

sampled from these block groups in proportion to population size. A total of 4,634 residential

addresses were successfully screened. We then drew stratified random samples of individuals 25

to 65 years of age, such that half were African American and half non-Hispanic white, with

roughly equal numbers of males and females. Although only a 60 percent interviewing success

rate was achieved, analyses have been weighted for the probability of non-contact during the

household screening phase and non-response during the interviewing phase. Post-stratification

weights were also incorporated into the final design weight to allow generalizability of our

findings to the surrounding community. For additional information on sampling and weights see

Turner, Brown, and Hale (in press).

Information on health status and life experiences was collected during three-hour,

computer-assisted in-person interviews. All participants signed informed consent forms prior to

the interview. During the interview, instructions regarding the 12-hour urine sample and the next

morning’s visit by the clinician were also provided. Clinicians arrived before breakfast, collected

the urine sample, drew blood samples, measured blood pressure three times spaced two minutes

apart, and took measures of the hip, waist, height, and weight. Information was also obtained on

medications, including those for blood pressure and high cholesterol. Virtually all participants

agreed to provide biomarker data, with fewer than two percent refusing the clinician visit. All

study-related procedures received ethical approval from the Institutional Review Board.

Measures

Adult Health

As previously noted, adult health is alternatively estimated in terms of three distinct

research definitions.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

12

Doctor Diagnosed Illness. Consistent with practices found in the literature (e.g. Hayward

et al., 2000) we obtained reports on the lifetime occurrence of 27 diseases and disorders. Here we

consider nine of these disorders that were judged to be serious, potentially fatal, and likely to

come to the attention of doctors: diabetes, heart disease, kidney problems, stroke, liver problems,

pancreatitis, tuberculosis, intestinal bleeding, high cholesterol. The count is limited to those

endorsed disorders that included a “yes” response to “was this health problem diagnosed by a

physician?”

Self-Rated Health. A four-item measure was employed (alpha = .75): a) “you seem to get

sick a little easier than other people,” b) “you are as healthy as anyone you know” (reverse

coded),” c) “you expect your health to get worse,” and d) “in general, your health is excellent”

(reverse coded). Participants responded across five categories ranging from (1) “definitely true” to

(5) “definitely false”. Self-rated health is the summed score across the four items; higher scores

indicate better self-rated health.

Allostatic Load (AL). Two categories of biomarkers are used to derive estimates of

allostatic load: primary mediators involving substances released by the body in response to stress,

and a set of secondary mediators generated from the effects of primary mediators. Following the

work and recommendations of the Allostatic Load Working Group of the MacArthur Research

Network, this study employed the primary mediators of epinephrine, norepinephrine, cortisol, and

dehydroepiandrosterone sulfate (DHEA-S) and the secondary mediators of systolic and diastolic

blood pressures, total cholesterol, high density lipids (HDL), glycated hemoglobin, and waist to

hip ratio (Seeman et al., 1997). Blood pressure readings are the average of three assessments

spaced two minutes apart. In the results to be presented, individuals who reported taking

medication for cholesterol are counted as having high total cholesterol and those taking blood

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

13

pressure medication are counted as having high systolic and high diastolic blood pressure. Repeat

analyses in which medications for cholesterol and blood pressure were not considered yielded

closely similar results.

Continuing the practice of the MacArthur research network, allostatic load is estimated by

a count of the number of dimensions with scores falling above the 75th percentile, except for

HDL and DHEA-S where the lowest quartile corresponds to highest risk. To adjust for the small

number of cases where data were not available on all 10 dimensions, the number of high-risk

dimension scores was divided by the number of dimensions with valid data.

Childhood Adversity

Early adversity is alternatively indexed in terms of exposures to traumatic events and level

of socioeconomic deprivation (childhood SEP).

Childhood Trauma. Major and potentially traumatic events were assessed utilizing a

slightly expanded list previously employed to evaluate the risk significance of early trauma for

psychiatric and substance disorders (Lloyd & Turner, 2008; Turner & Lloyd, 1995, 2003, 2004).

In the analyses presented, childhood is defined as including ages 15 and younger. The 32

potentially traumatic events judged to be of possible relevance to individuals 15 and under are

shown in Appendix A.

We employed a Life History Calendar (LHC) based on that developed by Freedman et al.

(1988) as an aid in achieving accurate recall of the timing of significant life course experiences.

This calendar traced several categories of experience. The first three involved a process in which

study participants described divisions of their lives in terms of places of residence (state, city,

street as appropriate), especially important people at the time (teachers, best friends, etc), and

landmark events (graduations, religious celebrations, births of siblings, wedding, divorce). These

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

14

dimensions are developed early in the interview one at a time, each building on the information

already in hand. The calendar is later employed for establishing the age of occurrence of each

reported major and potentially traumatic event. This is done by utilizing all information available

from scanning both upward and across the LHC. Thus, a reported trauma would be placed on the

calendar in relation to other traumatic events, if any ,teachers, best friends or significant others at

the time, noted landmark or transition events and place of residence. Total exposure is estimated

by the number of different events experienced through age 15.

Childhood SEP. Study participants were questioned about the family member who

provided the major financial support of their family “most of the time while you were growing

up.” Information on the educational attainment and occupation (in terms of job title and type of

work involved) was obtained. The occupational reports were coded using the Nam and Boyd

(2004) occupational status scores. They were independently coded twice with disagreements

reconciled in conference. The third dimension utilized to estimate childhood SEP was responses

to a question about the family‘s “financial situation most of the time while you were growing up.”

This dimension, in addition to the education and occupation of the main providing family

member, was standardized. Each of the standardized scores were then summed and the result

divided by the number of dimensions on which data were available. These scores were then re-

standardized. Supplementary analyses in which these components were disaggregated are also

reported.

Adult Stress Exposure

To provide a more stringent test of the possibility that the early adversity-adult health

relationship arises largely or wholly from their joint association with adult stress exposure, we

sought to achieve a more comprehensive assessment of exposure differences. Four dimensions of

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

15

stressful experience were considered: recent life events, chronic stress, lifetime major and

potentially traumatic events, and discrimination stress.

Recent life events (those occurring over the 12 months preceding the interview) were

assessed using a 33-item checklist of negative events previously employed in several studies

(Avison & Turner, 1988; Turner & Avison, 1989, 1992; Turner, Sorenson, & Turner, 2000;

Turner, Wheaton, & Lloyd, 1995). The items address such occurrences as trouble with the law,

serious illness, beaten up or physically attacked, fired or laid off, and a romantic relationship

ended.

Chronic stress. Our chronic stress measure was composed of 36 items developed using the

logic and items of Wheaton‘s (1991, 1994) measure as a starting point (see Turner & Avison,

2003; Turner et al., 1995 for a listing of these items). Items employed address life domains

including employment (e.g. “you don’t get paid enough for what you do,” “your work is boring

and repetitive”), relationships (e.g. “you have a lot of conflict with your partner,” “your partner is

not committed enough to your relationship”), and health concerns (e.g. “you have a parent, child

or partner who is in very bad health and may die,” “you take care of an aging or ill family member

or friend”). Participants were asked whether each statement is true for them at this time and, if

yes, how long it has been true. In the analyses to be presented, the measures of both recent life

events and chronic stress are simple counts of the number reported.

Lifetime exposure to major and potentially traumatic events. Lifetime exposure to major

and potentially traumatic events (major adversities) was measured with a relatively

comprehensive set of 32 questions. These questions had been used previously within research on

the significance of cumulative adversity for the subsequent onset of drug dependence, alcohol

dependence and psychiatric disorder (Lloyd & Turner, 2008; Turner & Lloyd, 2003, 2004).

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

16

Participants were asked whether each event had ever occurred, how many times, and their age at

first and last occurrence (see Appendix A for a listing). In order to increase confidence in our

assumed temporal ordering of stress exposure and health status and avoid possible confounding

between this adult dimension and our measure of childhood trauma, only events reported as

occurring between age 21 and two years prior to interview are counted.

Discrimination stress. Discrimination stress was assessed in terms of day-to-day

experiences utilizing a measure developed by Williams et al. (1997; see also Kessler, Mickelson,

& Williams, 1999). This scale was developed to assess discrimination regardless of the perceived

reason for discrimination (e.g., race/ethnicity, sex, age, gender, socioeconomic status). It captures

experiences that, for most part, involve character assaults that may or may not lead to an

interference with advancing one‘s socioeconomic position (Kessler et al., 1999:212). Day-to-day

discrimination is estimated by the sum of Likert scores across the nine items, which include such

statements as “you are treated with less courtesy than other people,” “you receive worse service

than other people at restaurants or stores,” and “people act as if they think you are dishonest.” The

internal reliability for this scale is .85. In the results presented below, these four dimensions of

stress exposure are combined and equally weighted. However, we also consider findings where

they are considered separately in supplemental analyses.

Adult Socioeconomic Position (SEP)

As in the case of childhood SEP, three components were employed to estimate

participants’ own adult SEP: education, occupation, and financial situation, where participants’

financial situation is indexed by reported total household income. Education, occupation (coded

in accordance with the 2004 Nam-Boyd guide), and household income were each standardized

and summed, with the sum divided by the number of dimensions on which data were available.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

17

Analytic Strategy

Analyses are conducted in three stages. First, descriptive statistics (means, proportions and

ranges) of the study variables are presented in Table 1. The second stage uses a multivariate

regression nested model strategy to examine the extent to which adversity in both childhood and

adulthood individually and collectively shape adult health. Tables 2 through 4 show estimates of

the effects of these factors on doctor-diagnosed disorders, self-rated health, and allostatic load,

respectively. While Ordinary Least Squares (OLS) regression is employed for self-rated heath,

Poisson regression models are utilized to predict chronic conditions and allostatic load because

they are count measures. A similar nested model strategy is used for all three health outcomes:

Model 1 examines the impact of childhood SEP; Model 2 estimates the effect of childhood

trauma; Model 3 shows the joint consequences of childhood SEP and trauma; Model 4 builds on

Model 3 by adding adult SEP; Model 5 builds on Model 3 by adding adult stress; Model 6 is the

full model that includes all childhood and adult SEP and stress measures. We are able to examine

the extent to which adult SEP and stress explain the effects of childhood SEP and trauma by

comparing the estimates of their effects across models. Third, to examine whether the effects of

childhood adversity are conditioned by subsequent adult social context or experience, or vice

versa, interactions among childhood and adult adversity dimensions were estimated. As noted

above, all analyses used a standard weighting procedure that adjusted for the probability of

inclusion in the study and were conducted using STATA 14 software.

RESULTS

------------------------------------------Table 1 about here--------------------------------------

Table 1 presents demographics and mean scores for the three distinct research definitions

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

18

of current health status and for the predictor variables to be evaluated. The second section

provides information on both indexes of early adversity that have received considerable attention.

With respect to experiences of major and potentially traumatic events, we report the average

number of major and potentially traumatic events out of 32 experienced through age 15. This

“total childhood trauma” measure represents one of our two principal measures of childhood

adversity. However, in the light of prior findings on the early adversity-adult health linkage, we

also distinguish total childhood trauma from items reflecting parental maladaptation and those for

which parental involvement might be reasonably assumed (see Appendix A for a listing of these

items). The relative significance of these three contrasting estimates of childhood trauma is

considered below.

Our primary analyses assess the explanatory significance of total adult stress exposure,

while also noting any additional information provided by supplemental disaggregated analyses.

--------------------------------------Table 2, 3, 4---------------------------------------------------------

Tables 2, 3, and 4 address the fundamental question of whether the early adversity-adult

health relationship is confirmed in present data across the three alternative estimates of adult

health status. Models 1, 2, and 3 within each table provide the relevant information. Overall,

estimates indicate that childhood SEP and childhood trauma are statistically significant predictors

of the three health outcomes in the expected direction—childhood SEP is negatively associated

with poor health while childhood trauma is positively associated with poor health. Childhood

SEP and childhood trauma are each predictive of both number of doctor-diagnosed disorders and

self-rated health when considered individually and remain so in the context of the other measure

(Tables 2 and 3). While allostatic load is effectively predicted by childhood SEP, no association

with childhood trauma is observed (Table 4). Despite some inconsistency, analyses across three

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

19

distinct operational definitions of adult health status generally confirm the linkage between

childhood adversity and adult health.

To address the question of whether these associations are independent of subsequent

social context and experience, regression analyses were extended. Specifically, we evaluated the

extent to which all, or some portion, of this over-time association is explained by intervening

differences in stress exposure and/or socioeconomic context. Clearly, the answer to this specific

question is yes. Indeed, aside from age, the only consistent finding across the three sets of

analyses is that both adult socioeconomic position (SEP) and adult exposure to social stressors

represent significant independent predictors. Clearly, the same cannot be said with respect to

either measure of childhood adversity. With the clear exception of the linkage between childhood

SEP and allostatic load, coefficients for both dimensions of adversity are non-significant in the

final models, having been either over explained (more than 100 percent of the slope accounted

for) or reduced in magnitude by 80 percent or more.

Comparisons across the three sets of analyses make clear that the operationalization of

health status chosen matters in several important regards. With respect to the core question of

biological embedding and associated action implications, only the finding revealing the

independent predictive significance of parental SEP in relation to allostatic load is consistent with

the contention that effective intervention must include a focus on childhood. However, it is worth

noting that even in this instance a significant minority of the parental SEP-allostatic load

relationship is accounted for by intervening social experience.

Other contrasts across the three analyses include substantial differences in variance

accounted for, ranging from a low of 12 percent for self-rated health to a high of 21 percent in the

case of allostatic load. Also noteworthy is the fact that the three sets of analyses are inconsistent

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

20

regarding sex and race differences in health risk. While no basic sex or race difference is

observed in relation to either self-rated health or doctor-diagnosed disorders, males and African

Americans are at significantly higher risk as estimated by allostatic load. It should be noted that

supplementary analyses (not shown) considered the possibility of racial differences in these

processes by conducting within-race analyses. Results conformed very closely to those shown

above, allowing confidence that the reported findings apply both within and across the two target

populations of the overall study.

One of the remaining questions associated with these three sets of analyses is whether the

health consequences of adult adversity are conditioned by level of childhood adversity, or vice

versa. We evaluated this possibility by estimating the interactions between each of the childhood

estimates and each adult measure of adversity, one at a time, in the context of Model 5 of each

analysis. None of the four interaction terms where significant in the case of allostatic load. For

self-rated health, a significant interaction between childhood SEP and adult SEP (p<.01) suggests

that high adult SEP is more consequential in the context of low childhood SEP, or vice versa. A

single significant interaction was also observed in analyses predicting doctor-diagnosed disorders.

Level of adult SEP matters most in the context of high levels of childhood exposure to trauma

(p<.05), or vice versa. That these conditioning effects are observed for different outcomes and for

only one of the three estimates of adult health status advises interpretive caution. Overall, these

results suggest that most of the effects of childhood adversity and of adult adversity are additive

in nature.

As noted above, we hoped to advance our understanding of the early adversity-adult

health relationship through the utilization of multi-dimensional measures that may more fully

capture variations in childhood and adult social contexts and in childhood and adult exposure to

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

21

stressors. However, in each case, the elements of the aggregated measures were equally

weighted, leaving the possibility that some elements may be more important for adult health than

others. To evaluate this possibility, analyses (not shown) were repeated with the three components

of SEP and of SEP of origin and the four dimensions of adult stress exposure disaggregated.

Presumably, because the individual components of each measure tend to be moderately

correlated, very few such components made a significant independent contribution in predicting

one or more of the three adult health status dimensions. Not including demographics, only

parental occupational level and the stress dimensions of adult trauma and chronic stress

significantly predicted allostatic load when considered individually. For doctor-diagnosed

disorders, only adult household income and three dimensions of adult social stress (recent events,

chronic stressors, and day-to-day discrimination stress) contributed significantly and

independently, while for self-rated health only adult household income and day-to-day

discrimination stress were significant. These findings suggest that the multidimensional measures

of childhood and adult SEP and of adult exposure to social stress tend to better represent the core

constructs under consideration.

Also noted above, many prior findings on the early trauma-adult health linkage have

operationalized trauma in terms of parental maladaptation. Given this body of literature we

repeated the analyses shown in Tables 2, 3, and 4 where early trauma was assessed in terms of

physical and/or emotional abuse, and again where six additional experiences for which parental

involvement might be reasonably assumed were also considered (see appendix A for a listing of

these items). The results of both sets of analyses (not shown) closely matched those presented

above based on a count of different major and potentially traumatic events experienced through

the first 15 years of life. In every case, the measure of childhood trauma employed was fully

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

22

accounted for by adult stress exposure and/or adult SEP.

DISCUSSION

In the context of a substantial and growing body of evidence indicating that childhood

adversity has long-term consequences for adult health, the present research aimed to contribute

toward a better understanding of the mechanisms involved and associated implications for

intervention efforts. As suggested in the introduction, the question of whether the childhood

adversity-adult health linkage is principally due to an embedding at the molecular level or to

cumulative insult processes (e.g., chains of risk) does not appear to be resolvable through

community-based research because it is not possible to adjudicate the relative significance of

early biological compromise and social contexts and experiences across the life course.

Embedding or scarring during childhood may effect later health by limiting or influencing adult

social and economic achievement and/or one’s capacity to gain and maintain social support and to

avoid or effectively cope with social stressors. Future research that also considers variations in

social and personal resources may contribute toward the identification of promising intervention

targets but will leave the question of the origins of such resources unanswered. What can be said

with some confidence, however, is that whatever the effects on adult health that derive from

biological embedding and/or from early social context and experience, they are primarily indirect

effects through adult social context and exposures.

This research has several advantages. First, recognizing that findings may vary with

differing research definitions of adult health, our analyses addressed three distinct and widely

used measures. Second, few prior studies had examined the joint health consequences of stress

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

23

and SEP during both childhood and adulthood, and even fewer had assessed these variables in a

relatively comprehensive fashion while also evaluating the relative effects of individual

components of these measures. Consequently, the question of the extent to which the effects of

childhood trauma and socioeconomic disadvantage on adult health are accounted for by adult

stress exposure and SEP had remained largely unanswered. This study answers the call by Hamil-

Luker and O‘Rand (2007) for an examination of the joint health effects of stress and disadvantage

in both childhood and adulthood. Third, with several noteworthy exceptions (e.g. Umberson et al.,

2014) previous studies on the matter have tended to focus on either childhood trauma or

childhood socioeconomic disadvantage, leading to two somewhat separate strands of research

(Miller et al., 2011). As a result, compelling evidence was lacking on the question of joint or

independent effects and of the substitutability of alternative indexes of early adversity. This study

offers such evidence.

However, it must be noted that there are grounds for urging caution in considering these

important findings. While they are based on representative samples of whites and African

Americans, these samples were drawn from a moderately sized southern city. Thus, the extent to

which our findings can be generalized to other geographical areas is uncertain. Although

confidence in the time ordering of the predictor and dependent variables considered seems

reasonable, the data are cross sectional and rely on retrospective reports of early adverse

experiences and circumstances. The extent to which life calendar procedures minimized reliability

and validity concerns associated with such reports cannot be estimated. While the results

presented above certainly confirm recent research documenting the long arm of childhood

disadvantage for adult morbidity (O‘Rand & Hamil-Luker, 2005), as well as mortality (Hayward

& Gorman, 2004; Warner & Hayward, 2006) and functional limitations (Haas, 2008), they are

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

24

also more revealing with respect to the mechanisms through which the effects of childhood

trauma and SEP are transmitted. Specifically, with one important exception, the health

consequences of childhood stressors are shown to be largely or fully accounted for by adult stress

exposure and/or adult SEP. To our knowledge, the combined explanatory significance of adult

SEP and adult stress exposure had not previously been evaluated.

The starting point for this work was the body of evidence indicating that persistent health

disparities in morbidity have their origins in childhood adversity and the associated implication

that public health efforts focused on reducing childhood adversity is likely to constitute the most

effective illness prevention strategy (Shonkoff et al., 2009). The results of this study are

consistent with and support this profound contention, as does a large body of research on child

development. However, their implications for prevention and intervention science are less

obvious. The observation that childhood SEP predicts allostatic load independent of adult context

and experience is consistent with a conclusion that this form of adversity involves biological

embedding or scaring that may not be redeemable by more advantaged adult context and social

experience. It would follow from this that intervention during childhood may be indispensable for

the prevention of some problematic health outcomes. However, this particular finding is also

consistent with the widely held expectation that benign changes in social context and experience

are also likely to be influential in relation to allostatic load.

Overall, the results of this study encourage the conclusion that effective intervention is

possible throughout the life course. Specifically, they suggest that efforts to reduce stress

exposure in adulthood and facilitate intergenerational upward socioeconomic mobility might yield

significant population-level health benefits, independent of adverse experiences during childhood.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

25

REFERENCES

Aneshensel, C. S. (2005). Research in mental health: Social etiology versus social consequences. Journal of Health and Social Behavior, 46(3), 221-28.

Aneshensel, C. S., Rutter, C.M. & Lachenbruch, P.A. (1991). Social structure, stress, and mental health: Competing conceptual and analytic models. American Sociological Review, 56,166-78.

Avison, W. R. & Turner, R.J. (1988). Stressful life events and depressive symptoms: Disaggregating the effects of acute stressors and chronic strains. Journal of Health and Social Behavior, 253-264.

Benyamini, Y., & Idler, E.L. (1999). Community studies reporting association between self- rated health and mortality additional studies, 1995 to 1998. Research on

Aging, 21(3), 392-401. Bifulco, A. T., Brown, G. W., & Harris, T. O. (1987). Childhood loss of parent, lack of

adequate parental care and adult depression: A replication. Journal of Affective Disorders, 12(2), 115-128.

Blane, D. (1999). The Life Course, the Social Gradient, and Health. In M. Marmot & R.G. Wilkinson (Eds.), Social Determinants of Health (pp. 65-80). New York: Oxford University Press.

Brown, T. H., Angela, M. O., & Adkins, D. E. (2012). Race-Ethnicity and health trajectories tests of three hypotheses across multiple groups and health outcomes. Journal of Health and Social Behavior, 53(3), 359-377.

Browne, A., & Finkelhor, D. (1986). Impact of child sexual abuse: A review of the research. Psychological Bulletin, 99(1), 66.

Bryer, Jeffrey, Nelson, B., Miller, J.B., & Krol, P. (1995). Childhood sexual and physical abuse as factors in adult psychiatric illness. Child Abuse, Physical and Emotional Abuse and Neglect, 1(4), 102.

Burman, M.A., Stein, J.A., Golding, J.M., Siegel, J.M., Sorenson, S.B., Forsythe, A.B., & Telles, C.A. (1988). Sexual assault and mental disorders in adult psychiatric illness. American Journal of Psychiatry, 148, 55-61.

Cassel, J. (1974). Psychosocial processes and stress: Theoretical formulation. International Journal of Health Services, 4(3), 471-482.

Cassel, J. (1976). The contribution of the social environment to host resistance. American Journal of Epidemiology, 104(2), 107-123.

Cohen, S., Doyle, W. J., & Skoner, D. P. (1999). Psychological stress, cytokine production, and severity of upper respiratory illness. Psychosomatic Medicine, 61(2), 175-180.

Cohen, S., Hamrick, N., Rodriguez, M. S., Feldman, P. J., Rabin, B. S., & Manuck, S. B. (2002). Reactivity and vulnerability to stress-associated risk for upper respiratory illness. Psychosomatic Medicine, 64(2), 302-310.

David, R., & Collins Jr, J. (2007). Disparities in infant mortality: What‘s genetics got to do with it? American Journal of Public Health, 97, 1191.

Dowd, J. B. & Goldman, N. (2006). Do biomarkers of stress mediate the relation between socioeconomic status and health? Journal of Epidemiology and Community Health, 60, 633-39.

Escarce, J. J., Epstein, K.R., Colby, D.C., & Schwartz, J.S. (1993). Racial differences in the elderly‘s use of medical procedures and diagnostic tests. American Journal of Public

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

26

Health, 83(7), 948-54. Farmer, M. & Ferraro, K. (2005). Are racial disparities in health conditional on socioeconomic

status? Social Science and Medicine, 60, 191-204. Ferraro, K. F., Farmer, M. M., & Wybraniec, J. A. (1997). Health trajectories: Long-term

dynamics among black and white adults. Journal of health and Social Behavior, 38- 54.

Ferguson, J. A., Tierney, W.M., Westmoreland, G.R., Mamlin, L.A., Segar, D.S., Eckert, G.J., Zhou, X.H., Martin, D.K., & Weinberger, M. (1997). Examination of racial differences in management of cardiovascular disease. Journal of the American College of Cardiology, 30, 1707-13.

Fincher, C., Williams, J.E., MacLean, V., Allison, J.J., Kiefe, C.I., & Canto, J. (2004). Racial disparities in coronary heart disease: A sociological view of the medical literature on physician bias. Ethnicity and Disease, 14, 360–71.

Freedman, D., Thornton, A., Camburn, D., Alwin, D., & Young-DeMarco, L. (1988). The life history calendar, a technique for collecting retrospective data. Sociological Methodology, 18(1), 37-68.

Galobardes, B., Lynch, J.W., & Smith, G.W. (2008). Is the association between childhood socioeconomic circumstances and cause-specific mortality established? Update of a systematic review. Journal of Epidemiologic Community Health, 62, 387-390.

Galobardes, B., Lynch, J.W., & Smith, G.D. (2004). Childhood socioeconomic circumstances and cause-specific mortality in adulthood: Systematic review and interpretation. Epidemiology Review, 26, 7-21.

Galobardes, B., Lynch, J.W., & Smith, G.W. (2006). Childhood socioeconomic circumstances predict specific causes of death in adulthood: The Glasgow student cohort study. Journal of Epidemiological Community Health, 60, 527-529.

Geronimus, A. T. (1992). The weathering hypothesis and the health of African-American women and infants: Evidence and speculations. Ethnicity and Disease, 2, 207-21.

Geronimus, A. T., Bound, J., Waidman, T.A., Hillemeier, M.M. & Burns, P.B. (1996). Excess mortality among blacks and whites in the United States. New England Journal of Medicine, 335, 1552-58.

Geronimus, A. T., 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(5), 826-33.

Gibson, R. C. (1991). Race and the self-reported health of elderly persons. Journal of Gerontology, 46(5), S235-S242.

Grant, I., Brown, G. W., Harris, T., McDonald, W. I., Patterson, T., & Trimble, M. R. (1989). Severely threatening events and marked life difficulties preceding onset or exacerbation of multiple sclerosis. Journal of Neurology, Neurosurgery & Psychiatry, 52(1), 8-13.

Haas, S. (2007). The long-term effects of poor childhood health: An assessment and application of retrospective reports. Demography, 44(1), 113-135.

Haas, S. (2008). Trajectories of functional health: The long arm of childhood health and socioeconomic factors. Social Science and Medicine, 66, 849-861.

Haas, S. & Rohlfsen, L. (2010). Life course determinants of racial and ethnic disparities in functional health trajectories. Social Science and Medicine, 70, 240-250.

Hägglöf, B., Blom, L., Dahlquist, G., Lönnberg, G., & Sahlin, B. (1991). The Swedish

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

27

childhood diabetes study: Indications of severe psychological stress as a risk factor for type 1 (insulin-dependent) diabetes mellitus in childhood. Diabetologia, 34(8), 579- 583.

Hamil-Luker, J., & Angela, M. (2007). Gender differences in the link between childhood socioeconomic conditions and heart attack risk in adulthood. Demography 44(1), 137- 158.

Hardy, M. A., Acciai, F., & Reyes, A.M. (2014). How health conditions translate into self-ratings: A comparative study of older adults across Europe. Journal of Health and Social Behavior, 55(3), 320-342.

Harris, T., Creed, F., & Brugha, T. S. (1992). Stressful life events and Graves' disease. The British Journal of Psychiatry, 161(4), 535-541.

Hayward, M. & Gorman, B. (2004). The long arm of childhood: The influence of early-life social conditions on men‘s mortality. Demography, 41(1), 87-107.

Hayward, M., Miles, T., Crimmins, E., & Yang, Y. (2000). The significance of socioeconomic status in explaining the racial gap in chronic conditions. American Sociological Review, 65(6), 910-930.

House, J. S., Lepkowski, J. M., Kinney, A. M., Mero, R. P., Kessler, R. C., & Herzog, A. R. (1994). The social stratification of aging and health. Journal of Health and Social Behavior, 213-234.

Idler, E. L. (1993). Age differences in self–assessments of health: Age changes, cohort differences, or survivorship. Journal of Gerontology, 48(6), S289-S300.

Idler, E. L., & Angel, R. J. (1990). Self-rated health and mortality in the NHANES-I Epidemiologic Follow-up Study. American Journal of Public Health, 80(4), 446-452.

Idler, E. L., & Kasl, S. V. (1995). Self-ratings of health, do they also predict change in functional ability? The Journals of Gerontology Series B, Psychological Sciences and Social Sciences, 50(6), S344-S353.

Masuda, M., & Holmes, T. H. (1978). Life events, perceptions and frequencies. Psychosomatic Medicine, 40(3), 236-261.

Jenkins, P. J., Harper, R. W., & Nestel, P. J. (1978). Severity of coronary atherosclerosis related to lipoprotein concentration. British Medical Journal, 2(6134), 388.

Johnson, R. J., & Wolinsky, F. D. (1994). Gender, race, and health: The structure of health status among older adults. The Gerontologist, 34(1), 24-35.

Kaplan, R.(1982). Health-Related quality of life measurement for evaluation research and health policy analysis. Health Psychology, 1(1),61.

Kaplan, G. A., & Keil, J.E. (1983). Socioeconomic Factors and Cardiovascular Disease, A Review of the Literature. Circulation, 88(4),1973-1998.

Kao, A.C., Green, D.C., Davis, N.A, Koplan, J.P., & Cleary, P.D. ( 1998a). Patients' trust in physicians, effects of choice, continuity, and payment method. Journals of General Internal Medicine, 13, 681-86.

Kao, Audiey C., Diane C. Green, Alan M. Zaslavsky, Jeffrey P. Koplan, and Paul D. Cleary. 1998b. The Relationship between Method of Physician Payment and Patient Trust.ǁ Journal of the American Medical Association 280,1708-14.

Karlamanga, A. S., Singer, B.H., & Seeman, T.E. (2006). Reduction in allostatic load in older adults is associated with lower all-cause mortality risk: MacArthur studies of successful aging. Psychosomatic Medicine, 68,500-7.

Karlsen, S., & Nazroo, J. Y. (2002). Relation between racial discrimination, social class, and

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

28

health among ethnic minority groups. American Journal of Public Health, 92(4), 624- 631.

Kessler, R. C., & Magee, W. J. (1994). Childhood family violence and adult recurrent depression. Journal of Health and Social Behavior, 13-27.

Kessler, R. C., Mickelson, K. D., & Williams, D. R. (1999). The prevalence, distribution, and Mental health correlates of perceived discrimination in the United States. Journal of Health

and Social Behavior, 208-230. Kivimaki, M., Leino-Arjas, P., Virtanen, M., Elovainio, M., Keltikangas-Jarvinen, L.,

Puttonen, S., Vartia, M., Brunner, E. & Vahtera, J. (2004). Work stress and incidence of newly diagnosed fibromyalgia, prospective cohort study. Journal of Psychosomatic Research, 57(5), 417-422.

Klabunde, C. N., Potosky, A.L., Harlan, L.C. & Kramer, B.S. (1998). Trends and Black/white differences in treatment for nonmetastatic prostate cancer. Medical Care, 36(9), 1337-48.

Krieger, N., & Sidney, S. (1996). Racial discrimination and blood pressure: The CARDIA study of young black and white adults. American Journal of Public Health, 86(10), 1370-1378.

Kuh, D. & Ben-Shlomo, Y. (2004). A Life-Course Approach to Chronic Disease Epidemiology. Oxford, England: Oxford University Press.

Kung, A. W. (1995). Life events, daily stresses and coping in patients with Graves' disease. Clinical endocrinology, 42(3), 303-308.

Leaverton, D. R., White, C. A., McCormick, C. R., Smith, P., & Sheikholislam, B. 1980. Parental Loss antecedent to childhood diabetes mellitus. Journal of the American Academy of Child Psychiatry, 19(4), 678-689.

Lloyd, D. A., & Turner, R. J. (2003). Cumulative adversity and posttraumatic stress disorder: Evidence from a diverse community sample of young adults. American Journal of Orthopsychiatry, 73(4), 381.

Lloyd, D. A., & Turner, R. J. (2008). Cumulative lifetime adversities and alcohol dependence in adolescence and young adulthood. Drug and Alcohol Dependence, 93(3), 217-226.

Lynch, J. & Smith, G.D. (2005). A life course approach to chronic disease epidemiology. Annual Review of Public Health, 26, 1-35.

McEwen, B. S. (1999). Protective and damaging effects of stress mediators: Central role of the brain. Progress in Brain Research, 122, 25-34.

McEwen, B. S. (2003). Mood disorders and medical illness: Mood disorders and allostatic load. Biological Psychiatry, 54, 200-7.

McEwen, B. S., & Stellar, E. (1993). Stress and the individual: Mechanisms leading to disease. Archives of Internal Medicine, 153(18), 2093-2101.

McEwen, B. S. & Seeman, T. (1999). Protective and damaging effects of mediators of stress: Elaborating and testing the concepts of allostasis and allostatic load. Annals of the New York Academy of Science, 896, 30-47.

McLeod, J. D. (1991). Childhood parental loss and adult depression. Journal of Health and Social Behavior, 205-220.

Miller, G., Chen, E. & Parker, K. (2011). Psychological stress in childhood and susceptibility to the chronic diseases of aging: Moving toward a model of behavioral and biological mechanisms. Psychological Bulletin, 137(6), 959-997.

Mooy, J. M., De Vries, H., Grootenhuis, P. A., Bouter, L. M., & Heine, R. J. (2000). Major

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

29

stressful life events in relation to prevalence of undetected type 2 diabetes, the Hoorn Study. Diabetes Care, 23(2), 197-201.

Nam, Charles B., & Boyd, M. (2004). Occupational status in 2000: Over a century of census-based measurement. Population Research and Policy Review, 23(4), 327-358.

Nerem, R. M., & Cornhill, J. F. (1980). The role of fluid mechanics in atherogenesis. Journal of Biomechanical Engineering, 102(3), 181-189.

Angela, M. O., & Hamil-Luker, J. (2005). Processes of cumulative adversity: Childhood disadvantage and increased risk of heart attack across the life course. The Journals of Gerontology Series B, Psychological Sciences and Social Sciences, 60(Special Issue 2), S117-S124.

Pearson, J. A. (2008). ‘Can‘t Buy Me Whiteness': New lessons from the Titanic on race, ethnicity and health. Du Bois Review, 5(1), 27-48.

Peterson, E. D., Shaw, L.K., DeLong, E.R., Pryor, D.B., Califf, R.M. & Mark, D.B. (1997). Racial variation in the use of coronary-revascularization procedures – Are the differences real? Do they matter? New England Journal of Medicine, 336, 480-486.

Pollitt, R, Rose, K.M. & Kaufman, J.S. (2005). Evaluating the evidence for models of life course socioeconomic factors and cardiovascular outcomes: A systematic review. BMC Public Health, 5, 7.

Repetti, R. L., Taylor, S.E., & Seeman, T. (2002). Risky families: Family social environments and the mental and physical health of offspring. Psychological Bulletin, 128(2), 330-366.

Rogers, M. P., Reich, P., Kelly, M. J., & Liang, M. H. (1980). Psychiatric Consultation among Hospitalized Arthritis Patients. General Hospital Psychiatry, 2(2), 89-94.

Ross, C. E., & Wu, C. L. (1995). The links between education and health. American Sociological Review, 719-745.

Rozanski, A., Blumenthal, J. A., & Kaplan, J. (1999). Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy.

Circulation, 99(16), 2192-2217. Saha, S., Arbelaez, J.J. & Cooper, L.A.( 2003). Patient-Physician relationships and racial

disparities in the quality of health care. American Journal of Public Health, 93(10), 1713-19.

Seeman, T. E., Singer, B.H., Rowe, J.W., Horwitz, R.I. & McEwen. B.S. ( 1997). Price of adaptation-allostatic load and its health consequences: MacArthur studies of successful aging. Archives of Internal Medicine, 157(19), 2259-68.

Seeman, T. E., Crimmins, E. Huang, M.H., Singer, B. Bucur, A., Gruenewald, T. Berkman, L.F. & Reuben, D.B. (2004) Cumulative biological risk and socio-economic differences in mortality: MacArthur studies of successful aging. Social Science & Medicine, 58(10), 1985-1997.

Shonkoff, J., Boyce, T. & McEwen, B. (2009). Neuroscience, molecular biology, and the childhood roots of health disparities. JAMA, 301(21), 2252-2259.

Smith, M. B. (1968). Competence and socialization. Socialization and Society, 270-320. Sonino, N., Girelli, M. E., Boscaro, M., Fallo, F., Busnardo, B., & Fava, G. A. (1993). Life

events in the pathogenesis of Graves' disease: A controlled study. Acta Endocrinologica, 128(4), 293-296.

Sternthal, M. J., Slopen, N., & Williams, D. R. (2011). Racial disparities in health. Du Bois Review, Social Science Research on Race, 8(01), 95-113.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

30

Stip, E., & Truelle, J. L. (1994). Organic personality syndrome in multiple sclerosis and effect of stress on recurrent attacks. Canadian Journal of Psychiatry. Revue Canadienne De Psychiatrie, 39(1), 27-33.

Taylor, J. & Turner, R.J. (2002). Perceived discrimination, social stress and depression in the transition to adulthood: Racial contrasts. Social Psychology Quarterly, 65, 213-225.

Thernlund, G. M., Dahlquist, G., Hansson, K., Ivarsson, S. A., Ludvigsson, J., Sjöblad, S., & Hägglöf, B. (1995). Psychological stress and the onset of IDDM in children: A case- control study. Diabetes Care, 18(10), 1323-1329.

Turner, R. J. (2003). Lifetime cumulative adversities and drug dependence: Racial/Ethnic contrasts. Addiction, 98, 305-15.

Turner, R. J. (2004). Stress burden and the lifetime incidence of psychiatric disorder in young adults: Racial and ethnic contrasts. Archives of General Psychiatry, 61(5), 481-488.

Turner, R. J. (2010). Understanding health disparities: The promise of the stress process model. In W. Avison, C.S. Aneshensel, S.Schieman, & B. Wheaton (Eds.), Advances in the conceptualization of the stress process (pp. 3-21). Springer.

Turner, R. J. ( 2013). Understanding health disparities: The relevance of the stress process model. Society and Mental Health.

Turner, R. J., Sorenson, A. & Turner, J.B. (2000). Social contingencies in mental health: A seven-year follow-up study of teenage mothers. Journal of Marriage and Family, 62(3), 777-791.

Turner, R. J. & Gil, A.G. (2002). Psychiatric and substance use disorders in south Florida: Racial/Ethnic and gender contrasts in a young adult cohort. Archives of General Psychiatry, 59, 43-50.

Turner, R. J. & Wheaton, B. (1995). Checklist measures of stressful life events. In L. Gordon, Cohen & R. Kessler (Eds.), Measuring Stress, A Guide for Health and Social Scientists (pp. 29-58) Oxford, UK: Oxford University Press.

Turner, R. J., Wheaton, B. & Lloyd, D.A. (1995). The epidemiology of social stress. American Sociological Review, 60, 104-25.

Turner, R. J., Brown, T. & Hale, W.B. (In press). Race socioeconomic position and physical health: A descriptive analysis. Journal of Health and Social Behavior.

Turner, R. J. & Lloyd, D.A. (1999). The stress process and the social distribution of depression. Journal of Health and Social Behavior, 40, 374-404.

Turner, R. J. & Lloyd, D.A. (1995). Lifetime traumas and mental health: The significance of cumulative adversity. Journal of Health and Social Behavior, 360-376.

Turner, R. J., Lloyd, D.A. & Taylor, J. (2006). Physical disability and mental health: An epidemiology of psychiatric and substance disorders. Journal of Rehabilitation Psychology, 51, 214-23.

Turner, R. J. & Avison, W. (1989). Gender and depression: Assessing exposure and vulnerability to life events in a chronically strained population. The Journal of Nervous and Mental Disease, 117(8), 443-455.

Turner, R. J. & Avison, W. (1992). Innovations in the measurement of life stress: Crisis theory and the significance of event resolution. Journal of Health and Social Behavior, 36-50.

Turner, R. J. & Avison, W.R. (2003). Status variations in stress exposure: Implications for the interpretation of research on race, socioeconomic status and gender. Journal of Health and Social Behavior, 44,499-505.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

31

Umberson, D., Williams, K., Thomas, P. A., Liu, H., & Thomeer, M. B. (2014). Race, gender, and chains of disadvantage childhood adversity, social relationships, and health.” Journal of Health and Social Behavior, 55(1), 20-38.

Vitaliano, P. P., Scanlan, J. M., Zhang, J., Savage, M. V., Hirsch, I. B., & Siegler, I. C. (2002). A path model of chronic stress, the metabolic syndrome, and coronary heart disease. Psychosomatic Medicine, 64(3), 418-435.

Warner, D. & Brown, T. (2011). Understanding how race/ethnicity and gender define age-trajectories of disability: An intersectionality approach.” Social Science and Medicine, 72, 1236-1248.

Warner, D. F. & Hayward, M.D. (2006). Early life origins of the race gap in men‘s mortality. Journal of Health and Social Behavior, 47, 209-226.

Warren, S., Greenhill, S., & Warren, K. G. (1982). Emotional stress and the development of multiple sclerosis: Case-control evidence of a relationship. Journal of Chronic Diseases, 35(11), 821-831.

Weissman, M. M., Gammon, G. D., John, K., Merikangas, K. R., Warner, V., Prusoff, B. A., & Sholomskas, D. (1987). Children of depressed parents: Increased psychopathology and early onset of major depression. Archives of General Psychiatry, 44(10), 847-853.

Wheaton, B. (1991). The specification of chronic stress, models and measurement. Society for the Study of Social Problems Annual Meetings, Cincinnati,Ohio.

Wheaton, B. (1994). Sampling the Stress Universe. In Stress and Mental Health (pp. 77-114). Springer.

Widom, C. S., Weiler, B. L., & Cottler, L. B. (1999). Childhood victimization and drug abuse: A comparison of prospective and retrospective findings. Journal of Consulting and Clinical Psychology, 67(6), 867.

Winsa, B., Karlsson, A., Bergstrom, R., Adami, H.O., Gamstedt, A., Jansson, R., Adamson, U. & Dahlberg, P.A.(1991). Stressful life events and Graves' disease. The Lancet, 338 (8781), 1475-1479.

Williams, D. R., Yu, Y., Jackson, J.S. & Anderson, N.B. (1997). Racial differences in physical and mental health: Socioeconomic status, stress, and discrimination.” Journal of Health and Social Behavior, 2(3), 335-351.

Williams, D. R. & Mohammed, S.A. (2013). Racism and health I: Pathways and scientific evidence. American Behavioral Scientist, 0002764213487340.

Zautra, A. J., Burleson, M. H., Matt, K. S., Roth, S., & Burrows, L. (1994). Interpersonal Stress, Depression, and Disease Activity in Rheumatoid Arthritis and Osteoarthritis Patients. Health Psychology, 13(2), 139.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

N 1252Age 45.98Gender

Female 0.47Male 0.53

RaceBlack 0.47White 0.53

Health Outcomes Allostatic Load (0-10) 3.05Doctor Diagnosed Illness (0-9) 0.76Self-Rated Health (4-20) 15.14Early Life Adversity Childhood Trauma

Maladaptive Parental Trauma (0-2) 0.19Parental Involved Trauma (0-6) 0.98Non-Parental Trauma (0-24) 1.34Total Childhood Trauma (0-32) 2.51

Childhood SEP low 0.33moderate 0.34high 0.33

Adult Social FactorsAdult SEP

low 0.32moderate 0.33high 0.35

Adult Stress Exposure Adulthood Trauma (0-20) 3.85Recent Life Events (0-15) 2.10Chronic Stress (0-40) 11.28Day-to-day Discrimination (0-43) 18.03Total Adult Stress Exposure (standardized) 0.04

Table 1. Means and Proportions of Study Variables

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6Age 0.03*** 0.03*** 0.03*** 0.03*** 0.03*** 0.03***

(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)Gender (Male=1) -0.06 -0.07 -0.06 -0.05 -0.04 -0.03

(0.06) (0.06) (0.06) (0.06) (0.06) (0.06)Race (White=1) 0.05 0.001 0.05 0.13 0.05 0.12

(0.06) (0.06) (0.06) (0.07) (0.06) (0.07)Childhood SEP -0.11** -0.09* -0.04 -0.07 -0.03

(0.04) (0.04) (0.04) (0.04) (0.04)Childhood Trauma 0.03*** 0.02* 0.02* -0.01 -0.01

(0.01) (0.01) (0.01) (0.01) (0.01)Adult SEP -0.16*** -0.14**

(0.04) (0.04)Adult Stress 0.06*** 0.06***

(0.02) (0.02)Intercept -0.51*** -0.70*** -0.63*** -0.70*** -0.59*** -0 .65***

(0.11) (0.12) (0.12) (0.12) (0.11) (0.12)N 1252 1252 1252 1252 1252 1252

R2 0.13 0.13 0.14 0.15 0.15 0.16F 38.96 41.78 34.25 29.92 30.66 27.05*p<0.05; **p<0.01; ***p<0.001

Note: standard errors are included in parentheses

Table 2. Doctor Diagnosed Illness Regressed on Early Trauma, Childhood SEP, Adult SEP, and Adult Stress Exposure

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6Age -0.02* -0.04*** -0.03** -0.03*** -0.03*** -0.04***

(0.01) (0.01) (0.01) (0.01) (0.01) (0.01)Gender (Male=1) 0.29 0.32 0.30 0.21 0.15 0.09

(0.21) (0.21) (0.21) (0.21) (0.21) (0.20)Race (White=1) -0.11 0.19 -0.08 -0.66** -0.09 -0.60**

(0.23) (0.21) (0.23) (0.23) (0.22) (0.23)Childhood SEP 0.63*** 0.53*** 0.14 0.42** 0.09

(0.15) (0.15) (0.15) (0.15) (0.15)Childhood Trauma -0.14*** -0.11** -0.10** 0.10* 0.09*

(0.04) (0.04) (0.04) (0.04) (0.04)Adult SEP 1.10*** 0.97***

(0.14) (0.14)Adult Stress -0.39*** -0.35***

(0.05) (0.05)Intercept 16.00*** 17.01*** 16.60*** 17.05*** 16.32*** 16.75***

(0.43) (0.48) (0.49) (0.47) (0.48) (0.46)N 1252 1252 1252 1252 1252 1252

R2 0.03 0.03 0.04 0.08 0.09 0.12F 8.06 7.29 7.99 17.18 18.44 23.70*p<0.05; **p<0.01; ***p<0.001

Note: standard errors are included in parentheses

Table 3. Self-Rated Health Regressed on Early Trauma, Childhood SEP, Adult SEP, and Adult Stress Exposure

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6Age 0.05*** 0.06*** 0.05*** 0.05*** 0.05*** 0.05***

(0.01) (0.01) (0.01) (0.01) (0.01) (0.01)Gender (Male=1) 0.449*** 0.44*** 0.05*** 0.47*** 0.50*** 0.51***

(0.11) (0.11) (0.11) (0.11) (0.11) (0.11)Race (White=1) -0.72*** -0.86*** -0.73*** -0.61*** -0.72*** -0.63***

(0.12) (0.11) (0.12) (0.13) (0.12) (0.12)Childhood SEP -0.27*** -0.26*** -0.18* -0.22** -0.16*

(0.07) (0.08) (0.08) (0.07) (0.07)Childhood Trauma 0.03 0.01 0.01 -0.05* -0.05*

(0.02) (0.02) (0.02) (0.02) (0.02)Adult SEP -0.03** -0.18*

(0.08) (0.08)Adult Stress 0.13*** 0.12***

(0.03) (0.03)Intercept 0.87*** 0.07* 0.08** 0.71** 0.89*** 0.081***

(0.22) (0.25) (0.25) (0.25) (0.25) (0.25)N 1222 1222 1222 1222 1222 1222

R2 0.19 0.18 0.19 0.20 0.21 0.21F 69.09 67.25 55.68 47.74 49.53 43.10*p<0.05; **p<0.01; ***p<0.001

Note: standard errors are included in parentheses

Table 4. Allostatic Load Regressed on Early Trauma, Childhood SEP, Adult SEP, and Adult Stress Exposure

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Appendix A. Frequencies of Individual Major and Potentially Traumatic Events Experienced through Age 15 (N=1,252)

Event Frequency Maladaptive Parental Trauma Regularly emotionally abused by one of your caretakers 0.15 Ever physically abused by one of your caretakers 0.08 Parental Involved Trauma Ever sent away from home or kicked out of the house because you did something wrong 0.05 Ever abandoned by one or both parents 0.13 Ever live in an orphanage, a foster home, a group home, or were a ward of the state 0.03 Parents ever divorce or separate 0.30 Either parent drink or use drugs so often/regularly that it caused problems for the family 0.25 Witness your mother or another close female relative being regularly physically or emotionally abused 0.21 Other Trauma Someone close to you drink or use drugs so often/regularly that it caused problems for the family 0.15 Ever lose your home because of a natural disaster such as a flood or fire 0.02 Had a serious accident, injury, or illness that was life-threatening or caused long-term disability 0.03 Ever witnessed a serious accident or disasters where someone else was hurt very badly or killed 0.12 Have sexual intercourse when you didn't want to because someone forced you or threatened to harm you 0.07 Were you ever touched or made to touch someone else in a sexual way because they forced/threatened you 0.14 Ever physically abused or injured by a boyfriend/girlfriend 0.008 Ever physically abused or inured by someone else you knew 0.05 Ever been tortured 0.01 Been shot at/threatened with a gun but not injured (excluding military service) 0.05 Been shot with a gun or badly injured with another weapon (excluding military service) 0.009 Ever been chased but not caught when you thought you could really get hurt 0.07 Been physically assaulted or mugged 0.04 Ever kidnapped or held captive 0.01 Seen someone you know chased but not caught or threatened with serious harm 0.06 Seen someone get shot at or attacked with another weapon 0.06 Ever seen someone seriously injured by gunshot or some other weapon 0.04 Actually seen someone get killed by being shot, stabbed, or beaten 0.03 Been in a car crash in which someone was killed or badly injured 0.04 Ever been told that someone you knew had been shot but not killed 0.05 Ever been told that someone you knew had been killed with a gun or other weapon 0.06 Anyone else you knew died suddenly or been seriously hurt 0.09 Been told that someone you knew killed him or herself 0.09 Been told that someone you knew had been raped 0.04 *p<0.05, **p<0.01, ***p<0.001

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

HIGHLIGHTS

Measures of childhood adversity are predictive of adult physical health

The early adversity-adult health linkage is largely explained by adult experience

Biological embedding of early adversity cannot be discounted

Prioritizing interventions to reduce childhood adversity are supported

Findings also suggest the likely effectiveness of adult-based interventions