DOI: 10.1542/peds.2013-3415; originally published online June 16, 2014;Pediatrics

A. Shirtcliff and Katherine P. TheallStacy S. Drury, Emily Mabile, Zoë H. Brett, Kyle Esteves, Edward Jones, Elizabeth

The Association of Telomere Length With Family Violence and Disruption  

  http://pediatrics.aappublications.org/content/early/2014/06/10/peds.2013-3415

located on the World Wide Web at: The online version of this article, along with updated information and services, is

 

of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2014 by the American Academy published, and trademarked by the American Academy of Pediatrics, 141 Northwest Pointpublication, it has been published continuously since 1948. PEDIATRICS is owned, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

at Tulane Univ on June 17, 2014pediatrics.aappublications.orgDownloaded from at Tulane Univ on June 17, 2014pediatrics.aappublications.orgDownloaded from

The Association of Telomere Length With Family Violenceand Disruption

WHAT’S KNOWN ON THIS SUBJECT: Poor health in children isassociated with exposure to family violence and disruption.Telomere length has been hypothesized to be a lasting biologicalindicator of exposure to early adversity and potentially predictiveof negative health outcomes throughout the life course.

WHAT THIS STUDY ADDS: Telomere length reflects exposure tofamily violence and disruption and may be an early indicator ofthe biological impact of early adversity. Children exposed tointerpersonal violence and family disruptions had significantlyshorter telomeres. Gender moderated these associations.

abstractBACKGROUND: To enhance the understanding of biological mecha-nisms connecting early adversity and negative health, we examinethe association between family interpersonal violence and disruptionand telomere length in youth. These specific exposures were selectedbecause of their established links with negative health consequencesacross the life-course.

METHODS: Children, age 5 to 15, were recruited from the greater NewOrleans area, and exposure to family disruption and violence wasassessed through caregiver report. Telomere length, from buccal cellDNA (buccal telomere length [bTL]), was determined by using mono-chrome multiplex quantitative real-time polymerase chain reaction.The association between bTL and adversity exposure was tested(n = 80).

RESULTS: Cumulative exposure to interpersonal violence and familydisruption was correlated with bTL. Controlling for other sociodemo-graphic factors, bTL was significantly shorter in children with higherexposure to family violence and disruption. Witnessing family violenceexerted a particularly potent impact. A significant gender interactionwas found (b = 20.0086, SE = 0.0031, z test= 22.79, P = .0053) andanalysis revealed the effect only in girls.

CONCLUSIONS: bTL is a molecular biomarker of adversity and allo-static load that is detectable in childhood. The present results extendprevious studies by demonstrating that telomeres are sensitive to ad-versity within the overarching family domain. These findings suggestthat the family ecology may be an important target for interventions toreduce the biological impact of adversity in the lives of children.Pediatrics 2014;134:e128–e137

AUTHORS: Stacy S. Drury, MD, PhD,a Emily Mabile, BA,b ZoëH. Brett, PhD,a Kyle Esteves, BA,a Edward Jones, BA,a

Elizabeth A. Shirtcliff, PhD,c and Katherine P. Theall, PhDb

aDepartment of Psychiatry, Tulane University School of Medicine,New Orleans, Louisiana; bDepartment of Global CommunityHealth, Tulane School of Public Health and Tropical Medicine, NewOrleans, Louisiana; and cDepartment of Psychology, University ofNew Orleans, New Orleans, Louisiana

KEY WORDStelomere length, toxic stress, interpersonal violence, familycontext

ABBREVIATIONSbTL—buccal telomere lengthCV—coefficient of variationPCR—polymerase chain reactionqPCR—quantitative PCRTL—telomere length

Dr Drury conceptualized and designed the study, drafted thefirst draft of the manuscript, and performed the initial statisticalanalysis; Ms Mabile supervised data collection, performed dataentry and cleaning, created tables and graphs, and criticallyreviewed the manuscript; Ms Brett performed all telomereassays, designed the quality control measures for the telomereassays, and critically reviewed the manuscript; Mr Estevessupervised and participated in all data collection, activelyrecruited study subjects, extracted DNA, performed qualitycontrol on DNA samples, and assisted with literature searchesand reviews; Mr Jones extracted DNA, performed quality controlon DNA samples, performed literature search, and assisted withdata cleaning; Dr Shirtcliff assisted in the conceptualization anddesign of the study, substantially contributed to the initialmanuscript, and reviewed and revised the manuscript;Dr Theall, with Dr Drury, conceptualized and designed the study,designed and performed all statistical analysis, contributed tothe initial draft of the manuscript, and reviewed and revisedsubsequent drafts; and all authors approved the finalmanuscript as submitted.

(Continued on last page)

e128 DRURY et al at Tulane Univ on June 17, 2014pediatrics.aappublications.orgDownloaded from

Fractured family contexts and exposureto family violence are all too frequentoccurrences for children.1 A 2009 studyrevealed that .60% of children underage 18 in the United States have beenexposed to direct or indirect violence inthe last year.2 One in every 28 childrenin the United States has a parent be-hind bars.3 Seven percent of the pop-ulation experiences the suicide ofa family member.4 Research, includingmeta-analytic studies, link poor mentaland physical health across the life-course with unstable family contextssuch as interpersonal family violence,5,6

parental suicide, and incarceration ofa family member.7–11 Although thenegative impact of these exposures isestablished, only 1 study to date hasexamined the association between anyof these exposures and telomerelength, an established biological in-dicator of cellular aging and, more re-cently, toxic stress.12,13

Exposure to early adversity increasesrisk for negative health outcomesacross the life course,14–16 yet it is un-clear whether specific experience orcumulative exposure is the most bi-ologically toxic. “Toxic stress” refers tothe biological changes to the stressresponse systems resulting from ex-posure to high levels of stressors thatoverwhelm a child’s ability to adapt,distinguishing it from lower levels ofstress, which can be beneficial, eg,“tolerable stress.” Challenges in de-fining and measuring toxic stress exist,and clear evidence of individual dif-ferences in children’s response andrecovery from even the most extremeearly experiences are found. As suchan enhanced understanding of howstress is, or is not, biologically em-bedded in youth, as well as biologicalmarkers that permit identification andtracking of these lasting effects arecritical. Toxic stress likely influenceschild health and development, partic-ularly risk for cardiovascular disease,

obesity, and mental illness, throughmultiple pathways. These pathwaysinclude the adoption of unhealthy lifestyles, decreased adaptive skills andcoping, and altered physiologic, cellu-lar, and immune stress responses. Inthe absence of the protective bufferinga child’s primary caregiver is expectedto provide, the negative trajectory ofthese pathways as a result of toxicstress is likely accentuated. Unfor-tunately the main factors associatedwith positive parenting and familystructure (parental sensitivity, sup-port, and availability) are often alsoaffected by stressors such as familyviolence, a known contributor to toxicstress.15 Implicit within this eco-bio-developmental framework is anemphasis on the family in relation tochild well-being, and a recognition thatfamily contexts may be both a potentialsource of stress (eg, family violence) aswell as a source of support (eg, stableadult caregiver).17 Given that parentalsupport is a known powerful bufferfrom toxic stress, disruption and dis-order within the family context may actsynergistically to negatively impactchildren.

The first straws of allostatic load, thecumulative wear and tear of intolerablestress on an individual’s biologicalsystems, likely begin during childhood,initiating a cascade of biologicalchanges to the stress response sys-tems,18 yet the observed changes maystill appear physiologically adaptive.19

In older individuals, where the physio-logic regulation of the stress responsesystems is less pliable and variable, thenegative trajectory of these changesbecomes more apparent. Although thedownstream health consequences ofexposure to high levels of early lifeadversity are readily observed inadults,20–22 validation of useful bio-markers within youth remains chal-lenging. As the biological memories ofearly adversity are putatively embed-

ded in gene by environment (g 3 e)interactions and epigenetic factors,molecular genetic approaches arelikely important methodologies for theidentification of novel indicators.15

Telomeres are repetitive DNA elementsthat cap the ends of chromosomes.Telomere length decline is a normativeconsequence of cellular division, aging,differentiation, and senescence. Accel-erated telomere shortening in bothadults and children has been associ-ated with stress-related physical dis-eases such as cardiovascular disease,obesity, and diabetes.23,24 Shortertelomere length in adults has also beenassociated with a history of childhoodmaltreatment and early adversity.25–27

The impact of early adversity on cellu-lar trajectories is already present inchildren as shorter telomeres, mea-sured in DNA extracted from blood,buccal swabs, and saliva, have beenassociated with environmental stressexposure (eg, community disorder andprenatal tobacco exposure),28,29 andpoor caregiving environments for thechild (eg, physical maltreatment, in-stitutional care, poverty), although onenegative study has been reported.13,30–32

Collectively these findings suggest thattelomere length has utility across thelife course, may link early adversity tonegative health outcomes, such asobesity and mental illness, even inyouth, and foreshadow increasedrisk.33

Although extreme deviations fromnormative caregiving (ie, institutional-ization or documented physical abuse)have been linked to shorter telomerelength in children, critically missingfrom previous research is whetherother indicators of family instabilityimpact cellular processes. Becausethere is substantial evidence linkingfamily violence, suicide, and in-carceration with negative health out-comes in youth, this study examinedwhether these family contextual

ARTICLE

PEDIATRICS Volume 134, Number 1, July 2014 e129 at Tulane Univ on June 17, 2014pediatrics.aappublications.orgDownloaded from

measures were associated with telo-mere length in children. Child genderwas a priori considered a potentialmoderator of the association as ourprevious studies, and those of others,have suggested gender moderates therelationship between early life adver-sity and telomere length.30

METHODS

Subjects

Children, age 5 to 15 years, wererecruited from greater New Orleans,Louisiana. Families were recruited byusing street outreach techniques, in-cluding ethnographic mapping andtargeted sampling,20,34 and throughtargeted schools in these communities.Recruitment neighborhoods were iden-tified by using the community identifi-cation process, a mapping method torecord epidemiologic indicators of theprevalence and incidence of commu-nity violence and other selected socialand health conditions.35 Interestedfamilies contacted the research site toschedule an appointment. Transpor-tation was provided and families werecompensated. This study was approvedby the Tulane University InstitutionalReview Board.

Data

Parental caregivers provided infor-mation about multiple levels of thechild’s social ecology (ie, householdand neighborhood) by using aninterview-assisted computer surveyadministered face-to-face at the re-search site (Nova Research, Bethesda,MD). Oral responses were recorded bytrained interviewers on a computer.Buccal swabs were collected from thechild for telomere length.

Measures

The primary outcome variable wasbuccal telomere length (bTL). DNA wascollected by using Isohelix SK1 buccalswabs (Cell Projects, Kent, United

Kingdom) and extracted by using theQIAamp DNA mini kit protocol (Qiagen,Valencia,CA).ConcentrationofextractedDNA was quantified with a Qubit dsDNABR assay kit (Invitrogen, Carlsbad, CA),purity of the DNA was determined byusing a NanoDrop 1000 spectropho-tometer (Thermo Fisher Scientific,Waltham, MA), and DNA integrity wasconfirmed by gel electrophoresis toensure highmolecularweight DNA. DNAwas stored at 280°C. The average rel-ative bTL was determined from thetelomere repeat copy number to singlegene (albumin) copy number (T/S) ratioby using an adapted monochrome mul-tiplex quantitative real-time polymerasechain reaction (PCR) and a BioRadCFX96.36 Ten microliters of DNA sample,containing ∼0.1 to 0.5 ng of DNA, wascombined with 15mL of PCR mixture, fora final volume of 25mL per reaction. ThePCR consisted of 0.753 Sybr Green I(Invitrogen), 13 Gene Amp Buffer II (Ap-plied Biosystems, Foster City, CA), 0.8 mMdeoxynucleotide triphosphates (dNTPs),10 mM magnesium chloride (MgCl2),3 mM Dithiothreitol (DTT), 1 M Betaine,2.5 U AmpliTaq Gold polymerase (Ap-plied Biosystems), 0.9 mM telg primer(ACACTAAGGTTTGGGTTTGGGTTTGGGTT-TGGGTTAGTGT), 0.9 mM telc primer(TGTTAGGTATCCCTATCCCTATCCCTATCCC-TATCCCTAACA), 0.6 mM albd2 primer(GCGGGCCCGCGTGGCGGAGCGAGGCCG-GAAAAGCATGGTCGCCTGT), and 0.6 mMalbu2 primer (GCCTCGCTCCGGGAGCGCCGCGCGGCCAAATGCTGCACAGAATCCTTG). The reaction proceeded for 1cycle at 95°C for 15 minutes, 2 cycles at94°C for 15 seconds, and 49°C for 1minute, 4 cycles at 94°C for 15 secondsand 59°C for 30 seconds, followed by 19cycles at 85°C for 15 seconds, and 73.5°Cfor 30 seconds, then 30 cycles at 94°Cfor 15 seconds and 84°C for 30 sec-onds. All samples were performed intriplicate, with a 7-point standardcurve (0.0313 ng to 2 ng) by usingpooled control DNA extracted frombuccal swabs. Triplicate plates were

repeated with all samples in a differentwell position. Thus, 6 replicates, of boththe single copy gene and the telomererepeat, were available for each in-dividual. PCR efficiency criteria fortelomere and albumin reactions were90% to 110%. Coefficients of variations(CVs) were calculated within each trip-licate (CV criteria# 10%) and betweenplates (CV criteria# 6%). Samples withunacceptably high CVs (10% intra- and6% interassay CV) were removed fromanalysis or repeated, resulting in a finalsample of 80. bTL ratio was determinedby the average of the triplicates fromboth plates. bTL was available on 80children. Children without bTL data didnot differ significantly (P . .05) fromchildren with bTL data on study mea-sures.

Our primary exposure was cumulativefamily instability defined bywitnessingfamily violence, family suicide, andincarceration. Parents were askedabout the child’s exposure to adverselife event by using questions extractedfrom the Preschool Age PsychiatricAssessment.36 Life events were cate-gorized as present or absent forwhether the child has (1) “… been ina situation where the child, the care-giver or someone close to the childcould have been hurt or killed?” (2)Experienced “A suicide of a familymember?” or (3) “a family memberincarcerated?” Cumulative family in-stability was classified as both a con-tinuous variable (0, 1, or 2+ events)and as high ($1 event) compared withlow (no events); analyses then exam-ined whether a particular type of ex-posure was a stronger predictor ofbTL.

Key Covariates

Demographics included child age inyears at the time DNA was collected,gender (boy or girl), maternal educa-tion as a marker of socioeconomicstatus, parental ageat child conception,

e130 DRURY et al at Tulane Univ on June 17, 2014pediatrics.aappublications.orgDownloaded from

and race.37 Race was self-reported andcategorized as African American (91%of the sample) or other (7%). Maternaleducation was classified as less thanhigh school, high school degree, somecollege, and a college or associatesdegree or more. Child’s age at DNAcollection was calculated from birth-date. Maternal and paternal age atchild’s conception was determinedfrom parent report. Missing paternalage (n = 11) scores were replaced withmean imputation.38

Data Analysis

Analyses were performed by using SASversion 9.2 (SAS Institute, Inc, Cary, NC).Bivariate analyses examined crudeassociations, including x2 or Fisher’sexact tests, t test or one-way analysis ofvariance. Sixty-two percent of enrolledfamilies had 1 child participate (range,1–5). To account for correlation be-tween siblings or children living in thesame household (intraclass correla-tion coefficient = 57%) and ensurecorrelated data did not inflate findings,generalized estimating equationsanalyses were employed by using anunstructured correlation structureusing PROC GENMOD. Generalized esti-mating equation permits one to takeinto account the most important clus-tering within groups. Three primarymultivariate analyses were conducted,including comparison of bTL acrosslevels of cumulative adversity, in-terpersonal violence, and family dis-ruption (ie, suicide/incarceration). Allanalyses controlled for child gender,age, maternal and paternal age atconception, race, and maternal edu-cation. Two-way interactions betweenrisk and child and maternal de-mographics examined potential mod-erating effects.

RESULTS

Table 1 shows respondent character-istics. Mean age of children was 10.2

years, and all analyses included childage at DNA collection as a covariate.bTL ranged from 1.11 to 3.00. Thirty-four children (43%) had experiencednone of our primary exposure events(eg, witnessing family violence, familysuicide, or incarceration of a familymember), 29 children (36%) had ex-perienced 1 event, and 17 children(21%) had experienced 2 or more.Forty-six children (57%) experiencedhigh cumulative adversity. Those withhigh (1+) levels of adverse experienceswere unexpectedly significantly morelikely to have mothers with higher ed-ucation and who were younger whenthe child was conceived. Parents’marital status was not correlated withbTL. Child BMI was not correlated withbTL, and controlling for BMI did notaffect results. As expected, child age

was significantly correlated with the

number of life events (r = 0.30, P, .01;

Table 1).

Mean bTL was significantly lower inchildrenwith high cumulative adversitycomparedwith those with less (t= 2.51,P = .01). Crude associations betweenbTL and overall cumulative risk, in-terpersonal violence, and family dis-ruption persisted after controlling forchild age, gender, race, maternal edu-cation, and parental age at conception(Table 2).

Association of bTL With CumulativeFamily Instability

We tested the hypothesis that cumula-tive exposure was associated with bTLin a dose dependent manner.15,38,39

Children experiencing the highest ad-versity had significantly shorter bTL

TABLE 1 Sample Characteristics of New Orleans Children Ages 5 to 15 y Overall and byCumulative Family Instability

Total (N = 80) Cumulative Family InstabilityLow (N = 34), High (N = 46)a,b

N (%) or mean (SD) N (%) or mean (SD) N (%) or mean (SD)

Gender of childBoy 41 (51) 21 (62) 20 (44)Girl 39 (49) 13 (38) 26 (57)

Race of childAfrican American 75 (94) 31 (91) 44 (96)Other 5 (6) 3 (9) 2 (4)

Age of child 10.2 (2.9) 9.9 (3.0) 10.4 (2.8)Parents’ marital statusSingle 66 (83) 30 (89) 36 (78)Married 14 (18) 4 (12) 10 (22)

Father’s age at conception 25.4 (7) 26.6 (8) 24.5 (6)Mother’s age at conceptionc 24.5 (6) 26.3 (7) 23.2 (5)Mother’s educational backgroundc

Grade school 20 (25) 9 (27) 11 (24)High school graduate or GED 19 (24) 12 (36) 7 (15)Some college, no degree 27 (34) 5 (15) 22 (48)College degree 13 (17) 7 (21) 6 (13)

Household monthly income$0–$999 30 (38) 11 (32) 19 (41)$1000–$1499 21 (26) 10 (29) 11 (24)$1500–$1999 13 (16) 3 (9) 10 (22)Over $2000 16 (20) 10 (29) 6 (13)

Average Telomere lengthc,d 1.8 (0.4) 1.9 (0.5) 1.7 (0.3)

GED, General Educational Development.a Values based on nonmissing data.b Cumulative family adversity is defined as parental reported child exposure to at least 1 out of the following 3 familyadversities: witnessing violence, incarceration of a family member, or family suicide.c P , .05 based on likelihood ratio x2 or Fisher’s exact or t test or Mann-Whitney U test where appropriate.d Average buccal cell telomere length as represented by the telomere repeat copy number to single gene (albumin) copynumber (T/S) ratio.

ARTICLE

PEDIATRICS Volume 134, Number 1, July 2014 e131 at Tulane Univ on June 17, 2014pediatrics.aappublications.orgDownloaded from

than unexposed children (b = 20.28,z = 22.21, P = .03). Further, childrenwith 1 exposure had significantlyshorter bTL than unexposed but in-termediate to the highest exposuregroup (b = 20.26, z =22.58, P = .01).Figure 1 shows mean bTL across levelsof adversity, with children who experi-enced no events exhibiting longer av-erage bTL than those who experienced1 or 2+ events (Fig 1).

We identified significant interactionbetween cumulative risk (b =20.0086,SE = 0.0031, z test = 22.79, P = .0053)and gender, and therefore stratifiedanalyses were performed. In girls, cu-mulative exposure remained signifi-cantly associated with bTL. Girls who

experienced 2+ exposures showeda half-unit decline in bTL comparedwith girls with no exposures (b =20.45, z = 22.26, P = .02), and girlswho experienced at least 1 adverseevent exhibited a similarly significantdecline (b =20.43, z =22.66, P = .01).In boys, cumulative exposure was nolonger significantly associated with bTL(Table 2, Fig 2), but significant covar-iates emerged. Analysis within boysrevealed that higher maternal educa-tional attainment was significantly as-sociated with longer bTL in thestratified analysis in boys. Further-more, bTL demonstrated the expectedinverse relation with age in boys(Fig 2).

Association of bTL With Witness toFamily Violence

Consistent with previous studies, wealso examined the impact of specificexposures. Children who witnessed fam-ily violence had significantly shorterbTL compared with unexposed children(b = 20.20, z = 22.21, P = .03), evenafter controlling for child age, gender,race, maternal education, and parentalage at child conception. As above, childgender moderated the association.Among girls, there remained a signifi-cant negative association between vi-olence exposure and bTL (b = 20.30,z = 22.21, P = .03). Among boys, expo-sure to violence was associated withshorter bTL but only reached trendlevel (b = 20.13, z = 21.76, P = .08).Among boys, child age, race, and ma-ternal education were significantly as-sociated with bTL. Specifically, olderage in boys was associated withshorter bTL and higher maternal edu-cation was associated with longer bTLin boys only. Maternal education andchild age were not associated with bTLwithin girls or within the total model(Table 2).

Association of bTL With FamilyDisruption

Children exposed to either or bothmarkers of family disruption (ie, familysuicide or incarceration) had shorterbTL than unexposed children at trendlevel (b =20.17, z =21.76, P = .08). Asabove, gender moderated this trend.Among girls, the association remainedmarginal (b =20.28, z =21.84, P= .07);however, in boys, family disruption wasnot significantly associated with bTL,but higher maternal education was as-sociatedwith longer bTL, and older boyshad significantly shorter bTL (Table 2).

DISCUSSION

This study extends previous telo-mere length research in children by

TABLE 2 Final Multivariate Models: Associations Between Familial Instability and AverageTelomere Length in Children (N = 80)

Overall Among Girls Among Boys

b (SE)* b (SE)* b (SE)*

Model A. Cumulative family instabilityand covariatesLow cumulative risk

(no adverse events)— — —

Medium cumulative risk(1 adverse event)

20.2563 (0.0992)** 20.4321 (0.1627)** 0.0183 (0.0817)

High cumulative risk (2 or moreadverse events)

20.2771 (0.1254)* 20.4453 (0.1968)* 20.1166 (0.0955)

Gender 0.1853 (0.0807)* — —

Age 20.0156 (0.0141) 20.0018 (0.0248) 20.0299 (0.0110)**Race 20.2833 (0.1280)* 20.5423 (0.1416)** 20.0952 (0.0908)Mother’s education 0.0108 (0.0312) 0.0087 (0.0502) 0.0540 (0.0225)*Mother’s age at conception 20.0031 (0.0096) 20.0124 (0.0188) 0.0137 (0.0094)Father’s age at conception 20.0013 (0.0070) 0.0051 (0.0146) 20.0079 (0.0065)

Model B. Witnessing violenceand covariatesWitnessing violence 20.2049 (0.0928)* 20.3020 (0.1368)* 20.1261 (0.0718)Gender 0.1734 (0.0872)* — —

Age 20.0125 (0.0152) 0.0089 (0.0293) 20.0273 (0.0103)**Race 20.1641 (0.1370) 20.3775 (0.1742)* 20.1591 (0.0493)**Mother’s education 20.0106 (0.0312) 20.0326 (0.0560) 0.0712 (0.0168)**Mother’s age at conception 0.0084 (0.0097) 20.0033 (0.0184) 0.0117 (0.0091)Father’s age at conception 20.0029 (0.0071) 20.0010 (0.0127) 20.0127 (0.0069)

Model C. Family disruptionand covariatesFamily disruption 20.1678 (0.0955) 20.2802 (0.1521) 20.0103 (0.0674)Gender 0.1618 (0.0809)* — —

Age 20.0178 (0.0143) 20.0041 (0.0258) 20.0362 (0.0101)**Race 20.2894 (0.1657) 20.8630 (0.2076)** 20.1131 (0.0868)Mother’s education 0.0045 (0.0319) 0.0059 (0.0538) 0.0602 (0.0219)**Mother’s age at conception 0.0054 (0.0093) 20.0099 (0.0173) 0.0129 (0.0095)Father’s age at conception 20.0017 (0.0073) 20.0003 (0.0146) 20.0066 (0.0064)

* P , .05; **P , .01.

e132 DRURY et al at Tulane Univ on June 17, 2014pediatrics.aappublications.orgDownloaded from

demonstrating that family instability(reflected by family incarceration,witnessed violence, and suicide) isassociated with bTL. Our findings fillan important intermediary gap in thegrowing literature where TL has beenassociated with both broad environ-mental/community-level adversity, aswell as adverse events experienced

directly by the child such as in-terpersonal violence, institutionalcare, and prenatal tobacco expo-sure.13,28–31,40 Even after controllingfor child age and maternal education,the link with bTL and family instabilitypersisted, particularly for witnessingfamily violence. Our findings are con-sistent with the existing TL studies in

adults and youth, as well as funda-mental principal of pediatrics: the de-velopment and health of children isintricately interwoven with the well-being of the entire family system.17,41

The impact of family instability on bTLwas significantly moderated by gender,such that the most robust effects wereobserved in girls. Gender differences

FIGURE 1Average TL in New Orleans children ages 5 to 15 by parental report of cumulative family instability. aAverage bTL as represented by the telomere repeat copynumber to single gene (albumin) copy number (T/S) ratio. bCumulative family instability is defined as parental reported child exposure to 3 family adversities:witnessing violence, incarceration of a family member, or family suicide.

FIGURE 2A, Average TL inNewOrleans childrenages 5 to 15byparental report of cumulative family instability by child gender. B, Average TL inNewOrleans childrenages 5to 15 by parental report of witnessing violence by child gender. aAverage bTL as represented by the telomere repeat copy number to single gene (albumin) copynumber (T/S) ratio. bCumulative family instability is defined as high or low. High refers to children with parental reported child exposure to at least 1 of 3family adversities: witnessing violence, incarceration of a family member, or family suicide. Low refers to no exposure. cWitnessing violence is defined byparental report.

ARTICLE

PEDIATRICS Volume 134, Number 1, July 2014 e133 at Tulane Univ on June 17, 2014pediatrics.aappublications.orgDownloaded from

have been reported in previous studiesof TL in youth, inmultiple studies of TL inadults, and in a range of studies directlyexamining the negative health con-sequences of early life stress, includingprenatal stress exposure.30,31,41 The linkbetween cumulative exposure and bTLdiffered in boys, with both maternaleducation and child age being in-fluential. Higher maternal educationalachievement was associated with lon-ger bTL in boys. Even in instable familycontexts, maternal education, or someother factor reflected in maternal ed-ucation, such as maternal resourcesand skills, was able, in part, to bufferthe impact of family instability withinboys.42 This finding is in agreementwith a previous study that revealed lowmaternal education associated withshorter child TL.43 Our current obser-vations deviate somewhat from pre-vious findings in that they raise thepossibility of a protective effect of ma-ternal education in families with in-stability. This would be consistent withShonkoff’s emphasis on the impor-tance of positive caregiving for chil-dren in adverse environments and thestudy by Asok et al31 in which parentalresponsiveness moderated the asso-ciation between high risk family envi-ronments and child TL.45–47 Recognizingthat stratification by gender results ineven smaller sample sizes, theseresults should be interpreted withcaution. However, evidence of gendereffects on telomere length dynamics,as well as gender differences in re-lation to the psychological and bi-ological impact of early adversity, doexist.47,48 Future studies specificallydesigned and sufficiently powered todefine gender effects that also capturedevelopmental considerations areneeded.

There are several strengths to thisstudy. First, this studywas a community-recruited sample. Although predictedto be high risk, this sample was not

selected for trauma exposure andtherefore is more generalizable to at-risk youth rather than specifically mal-treated individuals. Further, althougha minor percentage of other races wereincluded, the subjects were mostlyAfrican American, thus minimizing thepotentialof racial confounding.Althoughresults are reported for the entiresample, analysis within the AfricanAmerican subset was similar. Third, thisstudy controlled for parental age atconception.38 Together with previousstudies, our results indicate that directadverse events (ie, physical maltreat-ment), community level factors (com-munity disorder), and family contextsare all relevant predictors of cellularaging and stress as measured by telo-mere length. Our results further sug-gest that the family context mayfunction as both a stressor (ie, wit-nessing violence) and a buffer (mater-nal education).

Given the substantial body of researchlinking interpersonal and domestic vi-olence to negative health outcomesacross the life course, our findingssuggest that these experiences are in-deed biologically embedded and de-tectable early in the life course.49

Because telomere length has been as-sociated not only with early adversitybut also with the downstream negativehealth consequences associated withinterpersonal and domestic violence,differences in telomere length mayrepresent a precursor to these nega-tive health outcomes and a biologicalmarker with substantial utility fordirecting intervention and preventionefforts.

Despite its strengths, significant limi-tations including small sample size,self-reported measures of key expo-sures, and cross-sectional design, doexist.Wedidnot assess either theageatwhich the events occurred nor thefrequency of exposure. As such, therelative risk of earlier versus later ex-

posure, and the impact of repeatedexposures compared with singleexposures, cannot be determined. Asecond limitation is that we onlydetected an age related decrease in bTLin boys. Our inability to detect an age-related decline in bTL across bothgendersmay be due to thewide age anddevelopmental range of the sample andthe small sample size. A longitudinalstudy by Frenk and colleagues50 notedan apparent plateau in telomere lengthdecline beginning at age 4 until youngadulthood. This developmental effectmay further have limited our ability todetect age-related declines in telomerelength in the full sample. Althoughthere was no difference in age distri-bution between boys and girls, signifi-cant differences in pubertal statusexisted between boys and girls withgirls having more advanced pubertalstatus overall as would be expected.The potential of puberty influencing TLshould be evaluated in future studies.Larger studies, with more constrainedage ranges, as well as longitudinalstudies of children across devel-opment, are needed to disentanglepotential gender, pubertal status, andage associations with bTL in youth.

Limitations related to themeasurementof TL exist. In this study, TL was mea-sured by using the quantitative PCR(qPCR) method. Although the “goldstandard” of telomere measurement,the Southern blot, is preferable, inestablished molecular genetics labo-ratories the correlation between telo-mere restriction fragment length andqPCR results has been high.51 Recog-nizing that significant methodologicaldifferences in published studies utiliz-ing qPCR exist, with CV values revealedin other studies to range from 0.9% to28%, in this study bTLs were derivedfrom duplicate triplicate assays per-formed on separate plates for eachsubject. The average CV for triplicatesin this study was 3% indicating

e134 DRURY et al at Tulane Univ on June 17, 2014pediatrics.aappublications.orgDownloaded from

significant assay consistency andstringent methodology.

One limitation found across epigeneticstudies is the measurement of TL indifferent tissues. Multiple previousstudies of TL in children have beenreported, including 4 previous studiesutilizing buccal-derived TL,30,31,40,53 2 uti-lizing salivary derived TL,28,29 and overa dozen studies using peripheralblood.24,43,54 Our own pilot data suggesta high correlation of telomere lengthestimated concurrently from DNAextracted from buccal and blood (pairedt test: P = .94, not significantly different),and buccal and saliva samples (pairedt test: P = .96, not significantly different;Brett et al in preparation), consistentwith the cellular content overlap be-tween these tissues. As of yet, there islittle empirical evidence to suggest thatone peripheral source of DNA is themostaccurate estimator of TL.

CONCLUSIONS

Our findings add to the growing liter-ature linking early adversity and TL. Our

study is consistent with previous workthat demonstrated accelerated bTLdecline associated with exposure tomaltreatment, domestic violence, andfrequent bullying.40 Similarly, we foundthat although individual exposureswere predictive of bTL, cumulative ex-posure was the strongest predictor.Our results indicate that gender mod-erates the association between familyadversity and bTL and that, in boys,a maternal buffering effect may exist inparallel. Although girls appear directlyimpacted by family instability, the as-sociation in boys is more complicatedand may be related to other sociocul-tural factors not assessed in the cur-rent study. The finding of genderdifferences, even in childhood, relatedto the impact of family instability isconsistent with more recent findings inrelation to TL, the growing prenatalstress literature finding gender differ-ences even in infancy, and the estab-lished gender differences found acrosshealth outcomes in adults. The findingsof gender differences in our study,coupled with a recent study linking

testosterone levels and TL, suggest thatthe hypothalamic pituitary gonadalaxis may also be involved in the bi-ological embedding of early lifestress.55 Importantly, our results addto the accumulating evidence acrossa range of biomarkers that the lastingimpact of stress exposure is not limitedto adult outcomes and is identifiablemuch earlier in development.54 bTLmay represent an important tool tomonitor health and predict risk withvalidity from early developmental timepoints. Perhaps equally critical is that iffamily instability represents a sourceof biological increased risk, familystability may prove protective. Becausemost families include more than 1child, intervention efforts or pre-vention efforts targeting family stabil-ity may affect health outcomes inmultiple children. Moving forward,transdisciplinary studies that movebeyond identification, and towardmechanisms, are needed to most ef-fectively address persistent healthdisparities and the lasting conse-quences of early adversity.

REFERENCES

1. Sturge-Apple ML, Davies PT, Cicchetti D,Manning LG. Interparental violence, ma-ternal emotional unavailability and chil-dren’s cortisol functioning in familycontexts. Dev Psychol. 2012;48(1):237–249

2. Finkelhor D, Turner H, Ormrod R, Hamby SL.Violence, abuse, and crime exposure ina national sample of children and youth.Pediatrics. 2009;124(5):1411–1423

3. The Pew Charitable Trusts. Collateral Costs:Incarceration’s Effect on Economic Mobility.Washington, DC: The Pew Charitable Trusts;2010

4. Sorenson SB, Rutter CM. Transgenerationalpatterns of suicide attempt. J Consult ClinPsychol. 1991;59(6):861–866, discussion867–873

5. Davies PT, Sturge-Apple M, Martin J. Familydiscord and child health: an emotional

security formulation. In: Landale N, McHaleS, Booth A, eds. Family and Child Health.New York, NY: Springer; 2013:45–74

6. Blair C, Raver CC. Child development in thecontext of adversity: experiential canaliza-tion of brain and behavior. Am Psychol.2012;67(4):309–318

7. Murray J, Farrington DP, Sekol I. Children’santisocial behavior, mental health, druguse, and educational performance afterparental incarceration: a systematic reviewand meta-analysis. Psychol Bull. 2012;138(2):175–210

8. Roettger M, Boardman J. Parental in-carceration and gender-based risks forincreased body mass index: evidence fromthe National Longitudinal Study of Adoles-cent Health in the United States. Am JEpidemiol. 2011;175(7). Available at: http://

aje.oxfordjournals.org/content/early/2012/03/21/aje.kwr409. Accessed May 6, 2014

9. Lee RD, Fang X, Luo F. The impact of pa-rental incarceration on the physical andmental health of young adults. Pediatrics.2013;131(4). Available at: www.pediatrics.org/cgi/content/full/131/4/e1188

10. Wildeman C, Western B. Incarceration infragile families. Future Child. 2010;20(2):157–177

11. Geller A, Garfinkel I, Cooper CE, Mincy RB.Parental incarceration and child wellbeing:implications for urban families. Soc Sci Q.2009;90(5):1186–1202

12. Shalev I, Entringer S, Wadhwa PD, et al.Stress and telomere biology: a lifespan per-spective. Psychoneuroendocrinology. 2013;38(9):1835–1842

ARTICLE

PEDIATRICS Volume 134, Number 1, July 2014 e135 at Tulane Univ on June 17, 2014pediatrics.aappublications.orgDownloaded from

13. Shalev I. Early life stress and telomerelength: investigating the connection andpossible mechanisms: a critical survey ofthe evidence base, research methodologyand basic biology. Bioessays. 2012;34(11):943–952

14. Wegman HL, Stetler C. A meta-analytic re-view of the effects of childhood abuse onmedical outcomes in adulthood. PsychosomMed. 2009;71(8):805–812

15. Shonkoff JP, Garner AS; Committee onPsychosocial Aspects of Child and FamilyHealth; Committee on Early Childhood,Adoption, and Dependent Care; Section onDevelopmental and Behavioral Pediatrics.The lifelong effects of early childhood ad-versity and toxic stress. Pediatrics. 2012;129(1). Available at: www.pediatrics.org/cgi/content/full/129/1/e232

16. Felitti VJ, Anda RF, Nordenberg D, et al.Relationship of childhood abuse andhousehold dysfunction to many of theleading causes of death in adults. The Ad-verse Childhood Experiences (ACE) Study.Am J Prev Med. 1998;14(4):245–258

17. Boyce WT. The family is (still) the patient.Arch Pediatr Adolesc Med. 2009;163(8):768–770

18. Bush NR, Obradovi�c J, Adler N, Boyce WT.Kindergarten stressors and cumulativeadrenocortical activation: the “first straws”of allostatic load? Dev Psychopathol. 2011;23(4):1089–1106

19. Shirtcliff EA, Peres JC, Dismukes AR, Lee Y,Phan JM. Hormones: commentary. Ridingthe physiological roller coaster: adaptivesignificance of cortisol stress reactivity tosocial contexts. J Pers Disord. 2014;28(1):40–51

20. McEwen BS. Protective and damagingeffects of stress mediators. N Engl J Med.1998;338(3):171–179

21. Juster RP, McEwen BS, Lupien SJ. Allostaticload biomarkers of chronic stress andimpact on health and cognition. NeurosciBiobehav Rev. 2010;35(1):2–16

22. Danese A, McEwen BS. Adverse childhoodexperiences, allostasis, allostatic load, andage-related disease. Physiol Behav. 2012;106(1):29–39

23. Fitzpatrick AL, Kronmal RA, Gardner JP,et al. Leukocyte telomere length and car-diovascular disease in the cardiovascularhealth study. Am J Epidemiol. 2007;165(1):14–21

24. Buxton JL, Walters RG, Visvikis-Siest S,Meyre D, Froguel P, Blakemore AI. Child-hood obesity is associated with shorterleukocyte telomere length. J Clin Endo-crinol Metab. 2011;96(5):1500–1505

25. Kiecolt-Glaser JK, Gouin J-P, Weng NP,Malarkey WB, Beversdorf DQ, Glaser R.Childhood adversity heightens the impactof later-life caregiving stress on telomerelength and inflammation. Psychosom Med.2011;73(1):16–22

26. Kananen L, Surakka I, Pirkola S, et al.Childhood adversities are associated withshorter telomere length at adult age bothin individuals with an anxiety disorder andcontrols. PLoS ONE. 2010;5(5):e10826

27. Tyrka AR, Price LH, Kao HT, Porton B,Marsella SA, Carpenter LL. Childhood mal-treatment and telomere shortening: pre-liminary support for an effect of earlystress on cellular aging. Biol Psychiatry.2010;67(6):531–534

28. Theall KP, Brett ZH, Shirtcliff EA, Dunn EC,Drury SS. Neighborhood disorder andtelomeres: connecting children’s exposureto community level stress and cellular re-sponse. Soc Sci Med. 2013;85:50–58

29. Theall KP, McKasson S, Mabile E, DunawayLF, Drury SS. Early hits and long-termconsequences: tracking the lasting impactof prenatal smoke exposure on telomerelength in children. Am J Public Health. 2013;103(suppl 1):S133–S135

30. Drury SS, Theall K, Gleason MM, et al.Telomere length and early severe socialdeprivation: linking early adversity andcellular aging. Mol Psychiatry. 2012;17(7):719–727

31. Asok A, Bernard K, Roth TL, Rosen JB,Dozier M. Parental responsiveness moder-ates the association between early-lifestress and reduced telomere length. DevPsychopathol. 2013;25(3):577–585

32. Price LH, Kao H-T, Burgers DE, Carpenter LL,Tyrka AR. Telomeres and early-life stress:an overview. Biol Psychiatry. 2013;73(1):15–23

33. Watters JK, Biernacki P. Targeted sampling:options for the study of hidden pop-ulations. Soc Probl. 1989;36:416–430

34. Tashima N, Crain C, O’Reilly K, Elifson CS.The community identification (CID) process:a discovery model. Qual Health Res. 1996;6(1):23–48

35. Cawthon RM. Telomere length measure-ment by a novel monochrome multiplexquantitative PCR method. Nucleic Acids Res.2009;37(3):e21

36. Egger HL, Erkanli A, Keeler G, Potts E, WalterBK, Angold A. Test-Retest Reliability of thePreschool Age Psychiatric Assessment(PAPA). J Am Acad Child Adolesc Psychiatry.2006;45(5):538–549

37. Prescott J, Du M, Wong JYY, Han J, De Vivo I.Paternal age at birth is associated withoffspring leukocyte telomere length in the

nurses’ health study. Hum Reprod. 2012;27(12):3622–3631

38. Evans GW, Kim P. Childhood poverty andhealth: cumulative risk exposure andstress dysregulation. Psychol Sci. 2007;18(11):953–957

39. Rutter M, Garmezy N, Rutter M. Stress,coping and development: some issues. In:Garmezy N, Rutter M, eds. Stress, Copingand Development in Children. New York, NY:McGraw-Hill; 1983:1–41

40. Shalev I, Moffitt TE, Sugden K, et al. Expo-sure to violence during childhood is asso-ciated with telomere erosion from 5 to 10years of age: a longitudinal study. MolPsychiatry. 2013;18(5):576–581

41. Allmond BW, Buckman W, Gofman HF. TheFamily Is the Patient: An Approach to Be-havioral Pediatrics for the Clinician. AnnArbor, MI: Mosby; 1979

42. Steinberg L. Gallagher lecture. The familyat adolescence: transition and trans-formation. J Adolesc Health. 2000;27(3):170–178

43. Needham BL, Fernandez JR, Lin J, Epel ES,Blackburn EH. Socioeconomic status andcell aging in children. Soc Sci Med. 2012;74(12):1948–1951

44. Tibu F, Hill J, Sharp H, Marshall K, Glover V,Pickles A. Evidence for sex differences infetal programming of physiological stressreactivity in infancy [published onlineahead of print April 7, 2014]. Dev Psycho-pathol. doi.org/10.1017/S0954579414000194

45. Shonkoff JP. Science, policy, and practice:three cultures in search of a shared mis-sion. Child Dev. 2000;71(1):181–187

46. Shonkoff JP. Building a new bio-developmental framework to guide the fu-ture of early childhood policy. Child Dev.2010;81(1):357–367

47. Gardner M, Bann D, Wiley L, et al; HalcyonStudy Team. Gender and telomere length:systematic review and meta-analysis. ExpGerontol. 2014;51:15–27

48. Chaloner A, Greenwood-Van Meerveld B.Early life adversity as a risk factor forvisceral pain in later life: importance of sexdifferences. Front Neurosci. 2013;7:13

49. Finkelhor D, Turner HA, Hamby SL, OrmrodR. Polyvictimization: Children’s Exposure toMultiple Types of Violence, Crime, andAbuse. Washington, DC: US Department ofJustice, Office of Justice Programs, Officeof Juvenile Justice and Delinquency Pre-vention; 2011

50. Frenck RW Jr, Blackburn EH, Shannon KM.The rate of telomere sequence loss in hu-man leukocytes varies with age. Proc NatlAcad Sci USA. 1998;95(10):5607–5610

e136 DRURY et al at Tulane Univ on June 17, 2014pediatrics.aappublications.orgDownloaded from

51. Aviv A, Hunt SC, Lin J, Cao X, Kimura M,Blackburn EH. Impartial comparative anal-ysis of measurement of leukocyte telomerelength/DNA content by Southern blots andqPCR. Nucleic Acids Res. 2011;39(20):e134

52. Kroenke CH, Epel E, Adler NE, et al. Auto-nomic and adrenocortical reactivity andbuccal cell telomere length in kindergarten

children. Psychosom Med. 2011;73(7):533–540

53. Zhu H, Wang X, Gutin B, et al. Leukocytetelomere length in healthy Caucasian andAfrican-American adolescents: relation-ships with race, sex, adiposity, adipokines,and physical activity. J Pediatr. 2011;158(2):215–220

54. Theall KP, Drury SS, Shirtcliff EA. Cumulativeneighborhood risk of psychosocial stressand allostatic load in adolescents. Am JEpidemiol. 2012;176(suppl 7):S164–S174

55. Drury S, Shirtcliff EA, Shachet A, et al.Growing up or growing old? Cellular aginglinked with testosterone reactivity to stressin youth. Am J Med Sci. 2014; (in press)

(Continued from first page)

The content is solely the responsibility of the authors and does not necessarily represent the official views of the Eunice Kennedy Shriver National Institute of ChildHealth and Human Development or the National Institutes of Health.

www.pediatrics.org/cgi/doi/10.1542/peds.2013-3415

doi:10.1542/peds.2013-3415

Accepted for publication Apr 21, 2014

Address correspondence to Stacy S. Drury, MD, PhD, Department of Psychiatry and Behavioral Science, Tulane School of Medicine, 1430 Tulane Ave, #8055, NewOrleans, LA 70112. E-mail: sdrury@tulane.edu

PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).

Copyright © 2014 by the American Academy of Pediatrics

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

FUNDING: Supported by the National Institutes of Health (NIH; 1R01ES020447-01 to Dr Theall, and R21 MH094688-01 and 2K12HD043451-06 to Dr Drury) and TulaneUniversity (Dr Drury). The project described was supported by award number K12HD043451 from the Eunice Kennedy Shriver National Institute of Child Health andHuman Development. Funded by the National Institutes of Health (NIH).

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

ARTICLE

PEDIATRICS Volume 134, Number 1, July 2014 e137 at Tulane Univ on June 17, 2014pediatrics.aappublications.orgDownloaded from

DOI: 10.1542/peds.2013-3415; originally published online June 16, 2014;Pediatrics

A. Shirtcliff and Katherine P. TheallStacy S. Drury, Emily Mabile, Zoë H. Brett, Kyle Esteves, Edward Jones, Elizabeth

The Association of Telomere Length With Family Violence and Disruption  

ServicesUpdated Information &

/peds.2013-3415http://pediatrics.aappublications.org/content/early/2014/06/10including high resolution figures, can be found at:

Permissions & Licensing

tmlhttp://pediatrics.aappublications.org/site/misc/Permissions.xhtables) or in its entirety can be found online at: Information about reproducing this article in parts (figures,

Reprints http://pediatrics.aappublications.org/site/misc/reprints.xhtml

Information about ordering reprints can be found online:

rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Grove Village, Illinois, 60007. Copyright © 2014 by the American Academy of Pediatrics. All and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elkpublication, it has been published continuously since 1948. PEDIATRICS is owned, published, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

at Tulane Univ on June 17, 2014pediatrics.aappublications.orgDownloaded from