Neighborhood Social Conditions,Family Relationships, andChildhood AsthmaEdith Chen, PhD,a Robin Hayen, BA,a Van Le, BS,a Makeda K. Austin, BS,a Madeleine U. Shalowitz, MD,b Rachel E. Story, MD,c

Gregory E. Miller, PhDa

abstractBACKGROUND AND OBJECTIVES: Poor neighborhood conditions have established associations with poorerchild health, but little is known about protective factors that mitigate the effects of difficultneighborhood conditions. In this study, we tested if positive family relationships can buffer youthwho live in dangerous and/or disorderly neighborhoods from poor asthma outcomes.

METHODS: A total of 308 youths (aged 9–17) who were physician-diagnosed with asthma andreferred from community pediatricians and/or family practitioners participated in this cross-sectional study. Neighborhood conditions around families’ home addresses were coded by usingGoogle Street View images. Family relationship quality was determined via youth interviews.Clinical asthma outcomes (asthma symptoms, activity limitations, and forced expiratory volumein 1 second percentile), asthma management behaviors (family response to asthma symptomsand integration of asthma into daily life), and asthma-relevant immunologic processes(lymphocyte T helper 1 and T helper 2 cytokine production and sensitivity to glucocorticoidinhibition) were assessed via questionnaires, interviews, spirometry, and blood draws.

RESULTS: Significant interactions were found between neighborhood conditions and familyrelationship quality (b 5 j.11–.15j; P , .05). When neighborhood danger and/or disorder waslow, family relationships were not associated with asthma. When neighborhood danger and/ordisorder was high, better family relationship quality was associated with fewer asthmasymptoms, fewer activity limitations, and higher forced expiratory volume in 1 secondpercentile. Similar patterns emerged for asthma management behaviors. With immunologicmeasures, greater neighborhood danger and/or disorder was associated with greater T helper1 and T helper 2 cytokine production and reduced glucocorticoid sensitivity.

CONCLUSIONS: When youth live in dangerous and/or disorderly neighborhoods, high familyrelationship quality can buffer youth from poor asthma outcomes. Although families may notbe able to change their neighborhoods, they may nonetheless be able to facilitate betterasthma outcomes in their children through strong family relationships.

WHAT’S KNOWN ON THIS SUBJECT: Neighborhood conditions haveestablished associations with childhood asthma, but little is known aboutthe protective factors that can help buffer the effects of poor neighborhoodsocial environments on childhood asthma.

WHAT THIS STUDY ADDS: Under conditions of high neighborhood dangerand/or disorder, good family relationships were associated with fewerasthma symptoms, fewer activity limitations, and better pulmonaryfunctioning. Because neighborhoods are difficult to change, 1 route toimproving asthma outcomes may be through family relationships.

To cite: Chen E, Hayen R, Le V, et al. Neighborhood SocialConditions, Family Relationships, and Childhood Asthma.Pediatrics. 2019;144(2):e20183300

aDepartment of Psychology and Institute for Policy Research, Northwestern University, Evanston, Illinois; andcDepartment of Medicine and bCenter for Clinical Research Informatics, NorthShore University HealthSystem,Evanston, Illinois

Dr Chen conceptualized and designed the study, conducted data analyses, and drafted the initialmanuscript; Drs Miller, Shalowitz, and Story conceptualized and designed the study and reviewedand revised the manuscript; Ms Hayen and Ms Le coordinated and supervised data collection,contributed to the acquisition of data, and reviewed and revised the manuscript; and Ms Austincontributed to the acquisition of data and data interpretation and reviewed and revised themanuscript.

DOI: https://doi.org/10.1542/peds.2018-3300

Accepted for publication May 22, 2019

PEDIATRICS Volume 144, number 2, August 2019:e20183300 ARTICLE at Galter Health Sciences Library on December 19, 2019www.aappublications.org/newsDownloaded from

The characteristics of a neighborhood(both physical and social) caninfluence the health of those who livein that neighborhood.1,2 With respectto social conditions, living inneighborhoods with high residentialsegregation and economicdeprivation increases the risk ofadverse health outcomes, includingcardiovascular disease and overallmortality.3,4 Neighborhood violencealso increases the risk of asthma,obesity, and coronary heartdisease.5–8

Despite these risks, there are oftenobstacles that prevent families frommoving away from undesirableneighborhoods. As a result, they mustfind ways to adapt to the stressorstheir neighborhoods present. Little isknown about factors that facilitatesuch adaptation. Psychosocially,supportive family relationships maybe 1 such buffer, protectingindividuals from cardiometabolicrisks associated with other childhoodadversities (eg, low socioeconomicstatus, divorce, and abuse).9–11 Here,we tested if supportive familyrelationships serve a similar bufferingfunction in the context of childhoodasthma.

Asthma is the most prevalent chroniccondition in childhood, imposesa substantial economic burden on thecountry, and is associated withimpairment in multiple domains offunctioning.12 Asthma incidence ishigher in neighborhoods with adverseconditions (those with high violenceand disorder), and these conditionsare also associated with worseasthma control.13,14 We hypothesizedthat family relationships wouldinteract with neighborhood dangerand/or disorder to predict clinicaloutcomes (symptoms, activitylimitations, and pulmonaryfunctioning) in a sample of childrenwith asthma. We hypothesized 2reasons for such effects. Hypothesis 1was that positive family relationshipshelp youth maintain good asthmamanagement behaviors in the face of

difficult neighborhood conditions.Hypothesis 2 was that positive familyrelationships buffer the effects ofneighborhood conditions onbiological processes implicated instress and asthma.1 Asthma isa disease characterized byinflammation, constriction, andhypersensitivity of the airways. Thelper lymphocytes are key drivers ofthe pathology that initiates andmaintains these processes, andexcessive production of T helper 1(Th1) and T helper 2 (Th2)cytokines15 released by theselymphocytes are thought to beinvolved in multiple aspects ofasthma pathology.15–17 In addition,glucocorticoids play a major role inthe physiology and treatment ofasthma. Pharmacologic doses ofglucocorticoids inhibit Th1 and Th2cytokine production; however, thereis individual variability in sensitivityto these agents.18 We thus testedinteractions of neighborhood byfamily characteristics on asthmamanagement behaviors (familyresponses to symptoms andintegration of asthma into daily life)and on asthma-relevant immuneprocesses (Th1 and Th2 cytokineresponses to ex vivo stimulation andthe sensitivity of these responses toinhibition by glucocorticoids).

We used a novel approach toinvestigate neighborhood socialconditions. Authors of many previousstudies have relied on census-levelvariables or questionnaires tocharacterize neighborhoods. Censusdata capture broad structural aspectsof neighborhoods (eg, percentage ofpeople living below poverty) but notinformation about day-to-dayexperiences in neighborhoods.19

Questionnaires are by naturesubjective. Recent technologicaladvances allow researchers tocircumvent these problems andperform systematic observationsby taking a “virtual walk” withGoogle Street View to objectivelycharacterize neighborhoods (in terms

of the presence of graffiti ordeteriorated buildings, forexample).19,20

We hypothesized that when youthwith asthma live in more dangerousand disorderly neighborhoods, havingsupportive family relationships wouldbe associated with fewer asthmasymptoms and activity limitationsand better pulmonary function. Incontrast, when youth live in lessdangerous and/or disorderlyneighborhoods, family relationshipswould not be associated with asthmaoutcomes. We hypothesized thatsimilar interaction effects would befound for asthma managementbehaviors and asthma-relevantimmune processes.

METHODS

Participants

A total of 308 youths aged 9 to17 years old who were diagnosedwith asthma by a physician wererecruited from the Chicagoland areabetween 2013 and 2016 throughNorthShore University Health Systemand Erie Family Health Center, viareferrals from communitypediatricians and family practitioners.All patients had a current diagnosis ofasthma and had seen a physician forasthma within the past 24 months.Inclusion criteria were fluency inEnglish and no acute respiratoryillness at the visit. Exclusion criteriawere other chronic physical illnesses.Youth gave written assent, andparents provided written consent.This study was approved by theNorthwestern, NorthShore, and ErieInstitutional Review Boards.

Procedures

Participants completed interviews,questionnaires, spirometry, anda venous blood draw. Parents andyouth were each compensated $80.

Measures

See the Supplemental Information foradditional details.

2 CHEN et al at Galter Health Sciences Library on December 19, 2019www.aappublications.org/newsDownloaded from

Neighborhood Characteristics viaGoogle Street View

Trained raters completed virtualwalks through identified streetsegments using Google Earth Pro andGoogle Street View imagery andsystematic social observationmethods developed by Odgers et al.19

Virtual walks involved detailedinspection of streets, sidewalks,signage, and all buildings andadjacent yards or land. Neighborhoodcharacteristics recorded duringvirtual walks included signs ofphysical disorder (litter, run-downcars, small graffiti, large graffiti,scrubbed or painted-over graffiti, orother defaced property) and raters’global assessment of neighborhooddangerousness (based on thepresence of abandoned and/orboarded-up homes, vacant lots, barson windows and/or doors, and policecameras; in addition, the presence ofinstitutions such as churches orschools were considered mitigatingfactors). Interrater reliability for the 2scales ranged from 0.77 to 0.90.Because of their high correlation(r 5 0.65; P , .001), thedangerousness and disorder scaleswere standardized and averaged tocreate a single neighborhood dangerand/or disorder score.

Family Relationship Quality

Family relationship quality wasdetermined via interviews with youthusing the University of California LosAngeles Life Stress Interview.21,22

This interview is used to probe trust,support, and conflict in youth’srelationships with family members.Higher numbers reflected better-quality family relationships.

Asthma Clinical Outcomes

Pulmonary function was assessed inthe laboratory by using spirometry(MicroLoop; CareFusion, San Diego,CA) according to American ThoracicSociety guidelines.23 Forcedexpiratory volume in 1 second (FEV1)percentile was calculated on the basis

of child age, sex, ethnicity, andheight.24

Asthma activity limitations weremeasured by child report by using theActivity Limitations subscale of thePediatric Asthma Quality of LifeQuestionnaire.25 Higher scoresindicate greater activity limitations.

Parent report of child asthmasymptoms was queried with an itemquestion for parents: “How often hasyour child had a cough, wheeze,shortness of breath, or chest tightnessduring the past month?”26 Highernumbers indicated more asthmasymptoms.

Asthma Management Behaviors

The Family Asthma ManagementSystem Scale (FAMSS)27 isa semistructured interview that isused to query how families respondwhen they perceive breathingproblems in the child (familyresponse to symptoms and childresponse to symptoms) as well ashow well families balance managingasthma within their daily lives(balanced integration of asthma).Higher scores indicate betterresponses to symptoms and betterbalanced integration.

Asthma Immunologic Measures

Th1 and Th2 cytokine productionwere measured by ex vivo stimulatedperipheral blood mononuclear cells(PBMCs). A total of 0.5 3 106 PBMCswere isolated from venous blood bydensity-gradient centrifugation andincubated with 25 ng/mL of phorbol12-myristate 13-acetate (PMA) and1 mg/mL of ionomycin for 24 hours at37°C in 5% CO2.

28–30 Afterincubation, supernatants wereharvested and assayed in duplicatevia electrochemiluminescence ona Sector Imager 2400A (Meso ScaleDiscovery, Rockville, MD)31 for Th2(interleukin 4, interleukin 5,interleukin 10, and interleukin 13)and Th1 (interferon-g, interleukin 2)cytokines. Composite Th1 and Th2scores were derived by standardizing

each cytokine and then averagingvalues as described by Ehrlich et al.32

Glucocorticoid sensitivity wasmeasured by repeating the aboveprotocol, this time with 1.38 3 1026

M hydrocortisone added.33,34 At thisdose, cortisol suppresses productionof Th1 and Th2 cytokines, so highervalues reflect greater insensitivity toglucocorticoid inhibition.

Covariates

The covariates included child sex, age,ethnicity (white versus nonwhite),family income, asthma severity(based on the higher of symptomfrequency and medication use26), andwhether the child was using aninhaled corticosteroid (yes or no) anda b-agonist (yes or no for short actingor long acting).

Statistical Analyses

We conducted hierarchical multiple-regression analyses according toAiken and West,35 in which asthmaoutcomes were predicted from (1)covariates described above, (2) maineffects of neighborhood conditions(danger and/or disorder score) andfamily relationship quality (life stressinterview family relationship score),and (3) the interaction betweenneighborhood conditions and familyrelationships. Positive interactioncoefficients mean that at high levelsof 1 variable, there is a positiveassociation between the secondvariable and the outcome variable.Coefficients are presented asstandardized b coefficients, which canbe interpreted in terms of SD units;for example, a coefficient of .2 wouldindicate that for every 1 SD increasein the independent variable, thedependent variable increases by 0.2SDs. Predictor variables werecentered before analyses. Becauseneighborhood conditions and familyrelationships are continuousvariables, figures were created withestimated regression lines plotted at61 SD of these variables.

PEDIATRICS Volume 144, number 2, August 2019 3 at Galter Health Sciences Library on December 19, 2019www.aappublications.org/newsDownloaded from

RESULTS

Table 1 presents information aboutsample characteristics. Detailedresults are in Tables 2 through 4.

Main Effects of Family RelationshipQuality

Asthma Clinical Outcomes

There were no main effects of familyrelationship quality on asthma clinicaloutcomes.

Asthma Management Behaviors

Higher-quality family relationshipswere associated with betterfamily responses to asthmasymptoms (b 5 .18; P 5 .002) andalso with better balanced integrationof asthma into daily life (b 5 .24;P , .001).

Immunologic Measures

Family relationship quality was notrelated to immunologic measures.

Main Effects of Neighborhood Dangerand/or Disorder

Asthma Clinical Outcomes

Higher neighborhood danger and/ordisorder was associated with greaterasthma-related activity limitations(b 5 .18; P 5 .004).

Asthma Management Behaviors

Greater neighborhood danger and/ordisorder was associated with poorerfamily responses to asthmasymptoms (b 5 2.14; P 5 .027).

Immunologic Measures

Greater neighborhood danger and/ordisorder was associated with largerTh2 cytokine responses to PMA-ionomycin stimulation (b 5 .16;P5 .025) and with larger Th1 cytokineresponses (b 5 .14; P 5 .049).

Greater neighborhood danger and/ordisorder also was associated with lesssensitivity to glucocorticoids’inhibitory properties (ie, more Th2cytokine production despite theaddition of cortisol; b 5 .16;P 5 .026). The main effect ofneighborhood danger and/ordisorder for Th1 cytokine productionwith cortisol was b 5 .13; P 5 .08.

Interaction Effects: FamilyRelationships 3 NeighborhoodDanger and/or Disorder

Asthma Clinical Outcomes:ActivityLimitations

There was a significant interactionbetween family relationship qualityand neighborhood conditions(b 5 2.10; P 5 .048). Figure 1 revealsthat at lower levels of neighborhood

TABLE 1 Sample Characteristics

Mean SD %

Child age, y 12.99 2.50 —

SexMale — — 55

Child ethnicityWhite — — 50African American — — 26Hispanic — — 14Asian American — — 10

Family income 6.35 2.07 —

b-Agonist — — 97Inhaled corticosteroid — — 72Asthma severityMild intermittent — — 16Mild persistent — — 43Moderate — — 27Severe — — 14

Family relationship quality 2.06 0.79 —

FEV1, % 94.95 15.24 —

Activity limitations 3.34 1.30 —

Parent-reported asthma symptoms 1.31 0.59 —

Asthma managementFamily response 5.95 1.96 —

Child response 4.75 2.04 —

Balanced integration 5.54 1.68 —

Family income ranges from 1 to 9. The mean of 6.35 corresponds to the $75 000–$99 999 category. b-Agonist refers to longacting or short acting. Asthma severity ranges from 1 to 4. Family relationship quality ranges from 1 to 5. Activitylimitations range from 1 to 7. Parent-reported symptoms ranges from 1 to 4. Asthma management behaviors range from 1to 9. Values are not included for neighborhood danger and cytokine composites because they all have a mean of 0 and anSD close to 1 (given how scoring was done). —, not applicable.

TABLE 2 Regression Analyses of Neighborhood Danger and Family Relationship Quality PredictingAsthma Outcomes

Child-Reported ActivityLimitations, b

Parent-Reported AsthmaSymptoms, b

FEV1 %,b

Step 1Age 2.053 2.091 .059Sex 2.171** 2.066 2.076Ethnicity 2.171** 2.039 .106Income .033 .004 2.061Asthma severity .209** .523*** 2.056b-Agonist 2.003 .003 2.027ICS 2.188** 2.213*** 2.081

Step 2Neighborhood danger .180** 2.047 2.038Family relationship quality 2.105 2.027 .025

Step 3Neighborhood danger 3

family relationship2.109* 2.125* .129*

ICS, inhaled corticosteroid.* P , .05.** P , .01.*** P , .001.

4 CHEN et al at Galter Health Sciences Library on December 19, 2019www.aappublications.org/newsDownloaded from

danger and/or disorder, activitylimitations are low and do not vary byfamily relationship quality (b 5 .01;P 5 .94). At higher levels ofneighborhood danger and/or disorder,the higher the family relationshipquality, the fewer activity limitationsa child experiences (b 5 2.20;P 5 .008).

Asthma Clinical Outcomes: ParentReport of Asthma Symptoms

A significant interaction betweenneighborhood conditions and familyrelationship quality (b 5 2.12; P 5.016) indicated that at lower levels ofneighborhood danger and/ordisorder, parent-reported symptomsdo not vary by family relationshipquality (b 5 .10; P 5 .19). At higherlevels of neighborhood danger and/ordisorder, the higher the familyrelationship quality, the fewer parent-reported symptoms a child has (b 52.13; P 5 .046).

Asthma Clinical Outcomes: Spirometry

A significant interaction effect (b 5.13; P 5 .027) indicated that at lowerlevels of neighborhood danger and/ordisorder, FEV1 does not vary byfamily relationship quality (b52.11;P 5 .21). At higher levels ofneighborhood danger and/ordisorder, the association betweenfamily relationship quality and FEV1

was b 5 .13; P 5 .07.

Asthma Management Behaviors: FamilyResponse

There was a significant interactioneffect for family response to asthmasymptoms (b 5 .13; P 5 .025).Figure 2 reveals that at lower levelsof neighborhood danger and/ordisorder, family response to asthmasymptoms is good and does not varyby family relationship quality (b 5.06; P 5 .49). At higher levels ofneighborhood danger and/ordisorder, the higher the family

relationship quality, the bettera family’s response to asthmasymptoms (b 5 .29; P , .001).

Asthma Management Behaviors: ChildResponse

A significant interaction effect (b 5.11; P 5 .05) revealed that at lowerlevels of neighborhood danger and/ordisorder, child response to asthmasymptoms does not vary by familyrelationship quality (b 5 2.03;P 5 .74). At higher levels ofneighborhood danger and/or disorder,the higher the family relationshipquality, the better a child’s response tosymptoms (b 5 .18; P 5 .02).

Asthma Management Behaviors:Balanced Integration

There was an interaction effect forbalanced integration (b 5 .15; P 5.006). Figure 2 reveals that at lowerlevels of neighborhood danger and/ordisorder, balanced integration doesnot vary by family relationshipquality (b 5 .09; P 5 .26). At higherlevels of neighborhood danger and/ordisorder, the higher the familyrelationship quality, the morebalanced integration a family has(b 5 .36; P , .001).

Asthma Management Behaviors:Immunologic Measures

There were no interaction effectsbetween family relationship qualityand neighborhood conditions forimmunologic measures.

DISCUSSION

The current study is the first that weare aware of to leverage Google StreetView technology to better understandvariations in children’s asthmaoutcomes. Using systematicobservations, we found that dangerand disorder in children’sneighborhoods was associated withworse asthma outcomes acrossmultiple indicators. Additionally, wefound evidence suggesting thatpositive family relationships mayoverride these effects. Specifically,

FIGURE 1Interaction between neighborhood conditions and family relationship quality predicting asthmaactivity limitations, with higher scores indicating more activity limitations. The figure depicts esti-mated regression lines at 61 SD of neighborhood danger. Lower and higher family relationshipquality also refer to 61 SD.

PEDIATRICS Volume 144, number 2, August 2019 5 at Galter Health Sciences Library on December 19, 2019www.aappublications.org/newsDownloaded from

when neighborhood danger and/ordisorder is high, having higher-qualityfamily relationships is associated with

fewer parent-reported asthmasymptoms, fewer asthma activitylimitations, and better pulmonary

functioning. Youth exposed to highneighborhood danger and/ordisorder but with high-quality familyrelationships had asthma profilessimilar to youth in low danger and/ordisorder neighborhoods.

These findings are consistent withprevious research that hasdocumented links betweenneighborhood violence andasthma,5,13,14 and research that hasdocumented associations betweenneighborhood danger and otherhealth outcomes such as obesityand depression.36–39 In the currentstudy, we advance previousresearch in 2 ways. First, we movebeyond the main effects ofneighborhoods to examiningprotective factors. The 1 previousstudy in which Google Street Viewimages were used revealed themain effects of neighborhooddisorder and danger on children’santisocial behaviors.19 Here, weextend those findings to a physicalhealth outcome and find evidencethat positive family relationshipsbuffer youth from the health risksassociated with neighborhooddanger and/or disorder. Thisfinding is consistent with otherresearch on family relationships asstress buffers against healthproblems9–11,40–42 but extends thatwork to neighborhood conditions.

The second advance is showcasingthe utility of a new methodology forneighborhood and health research.Authors of many previous studiesrelied on participant reports ofneighborhood characteristics, whichare open to subjective biases. In thecurrent study, we address this byhaving coders rate neighborhoodconditions. What makes this possibleto do is the wealth of images availablethrough Google Street View, wherebya coder can take a walk througha person’s neighborhood virtually.This eliminates the expense of havingto send raters driving through theneighborhoods of participants.20 Inaddition, it is a more meaningful

FIGURE 2Interaction between neighborhood conditions and family relationship quality predicting asthmamanagement behaviors. A, How well families respond to their child’s asthma symptoms. B, How wellfamilies balance managing asthma within their daily lives. In both cases, higher scores indicatebetter asthma management. The figure depicts estimated regression lines at 61 SD of neighbor-hood danger. Lower and higher family relationship quality also refer to 61 SD.

6 CHEN et al at Galter Health Sciences Library on December 19, 2019www.aappublications.org/newsDownloaded from

indicator of neighborhoods thancensus variables. There are struggleswith census data in terms ofaccurately defining what representsa neighborhood to an individual andin terms of census variables notalways reflecting people’s livedexperiences in a neighborhood. Incontrast, Google Street View coding isfocused on a narrowly defined,precise area that reveals where

participants walk and what they seein their daily lives.

Through what processes wouldpositive family relationships bufferchildren from the effects of dangerousor disorderly neighborhoods? Oneexplanation is that when families livein difficult neighborhoods, positivefamily relationships may mean thatparents are making efforts to shield

their children by not allowing theneighborhood environment to disruptfamily life and routines. For childrenwith asthma, this may mean thatparents prioritize asthmamanagement and teach their childrenhow to stay focused on maintainingtheir health (eg, being aware ofsymptoms and responding to them ina timely manner), regardless of whatis going on in their neighborhood.

In contrast, it did not appear thatpositive family relationships bufferedphysiologic responses to stressbecause there were no interactionspredicting cytokine production.Rather, there were direct effects ofneighborhoods, such that worseconditions were associated withlarger Th1 and Th2 responses tomitogenic stimulation and lowerglucocorticoid sensitivity of Th2cytokines. These patterns suggestthat neighborhood socialcharacteristics may function asanother level of stressor linked tobiological processes in asthma,similar to stressors at the individualand family levels.33,43–47 It is possiblethat biological responses toneighborhood stressors may bebuffered by individual-levelcharacteristics such as coping48,49 orfunctional connectivity of brainnetworks involved in self-regulation.50

Strengths of the current study includethe systematic social observation ofneighborhoods using Google StreetView, the interview-based measuresof family relationships and asthmamanagement (considered goldstandards), and the battery ofimmune processes assessed in youth.Limitations include the fact thatcausality cannot be determined fromobservational data and the cross-sectional nature of this study design.In addition, Google Street Viewimages provide a 1-time snapshotthat may not accurately capturea neighborhood over time, and wewere limited by only having currentfamily addresses.

TABLE 3 Regression Analyses of Neighborhood Danger and Family Relationship Quality PredictingAsthma Management Behaviors

Family Response toAsthma Symptoms, b

Child Response toAsthma Symptoms, b

Integration ofAsthma ManagementInto Daily Life, b

Step 1Age 2.157** 2.040 2.005Sex 2.079 2.162** .152**

Ethnicity 2.051 .015 .197**

Income .091 .120* .087Asthma severity 2.014 2.061 2.232***

b-Agonist .100 .148* .059ICS .066 .065 .156*

Step 2Neighborhood danger 2.142* 2.102 2.112Family relationship

quality.185** .084 .235***

Step 3Neighborhood danger 3

family relationship.127* .110 .147**

ICS, inhaled corticosteroid.* P , .05.** P , .01.*** P , .001.

TABLE 4 Regression Analyses of Neighborhood Danger and Family Relationship Quality PredictingAsthma-Relevant Immunologic Outcomes

PMA-IonomycinTh1 Cytokines, b

PMA-IonomycinTh2 Cytokines, b

PMA-Ionomycinand Cortisol Th1Cytokines, b

PMA-Ionomycinand Cortisol Th2Cytokines, b

Step 1Age 2.130* .112 2.060 .087Sex 2.021 .042 2.004 2.027Ethnicity 2.118 2.161* 2.096 2.127Income .002 .052 .056 .069Asthma severity 2.034 2.005 2.053 2.004b-Agonist 2.005 2.006 2.015 2.010ICS .035 2.020 .098 .034

Step 2Neighborhood danger .139* .158* .128 .163*

Family relationshipquality

2.012 2.056 2.060 2.016

Step 3Neighborhood danger

3 familyrelationship

.042 2.036 .027 2.056

PMA-ionomycin is the incubation of PBMCs with 25 ng/mL of phorbol 12-myristate 13-acetate 1 1 mg/mL of ionomycin. ICS,inhaled corticosteroid.* P , .05.

PEDIATRICS Volume 144, number 2, August 2019 7 at Galter Health Sciences Library on December 19, 2019www.aappublications.org/newsDownloaded from

Implications of this study include theidea that if families have to live indangerous and/or difficultneighborhoods, there may be optionsfor minimizing the effects ofneighborhood conditions on theirchild’s asthma. In particular, strongfamily relationships may helppromote better asthma outcomes inyouth perhaps because high-qualityrelationships help youth figure outhow to manage their asthma despiteneighborhood stressors. In futurestudies, the efficacy of familyinterventions among youth withasthma who live in high-dangerneighborhoods could be tested. Thesetypes of interventions have beenfound to improve inflammatorycytokine profiles,51 cortisol levels,52

and outcomes in youth withdiabetes,53,54 so they might be

beneficial among youth with asthmaas well. These types of interventionsalso have the strongest effects onthose who are at highest risk atprogram entry,55–58 so they might beparticularly useful if targeted at youthliving in difficult neighborhoods.

CONCLUSIONS

In the current study, wedemonstrated that whenneighborhoods are characterized asdangerous and disorderly, better-quality family relationships areassociated with better asthmamanagement behaviors and betterasthma outcomes. These patterns areapparent across multiple clinicalindicators, including symptoms,activity limitations, and lung function.Together, the results suggest that

although families may not be able todo much to change theneighborhoods in which they live,they may nonetheless be able tofacilitate better asthma outcomes intheir children through strong familyrelationships.

ABBREVIATIONS

FAMSS: Family AsthmaManagement System Scale

FEV1: forced expiratory volume in1 second

PBMC: peripheral bloodmononuclear cell

PMA: phorbol 12-myristate 13-acetate

Th1: T helper 1Th2: T helper 2

Address correspondence to Edith Chen, PhD, Department of Psychology, Northwestern University, 2029 Sheridan Rd, Evanston, IL 60208. E-mail: edith.chen@

northwestern.edu

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

Copyright © 2019 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 National Institutes of Health grant R01 HL108723. 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.

REFERENCES

1. Diez Roux AV, Mair C. Neighborhoodsand health. Ann N Y Acad Sci. 2010;1186:125–145

2. Schreier HM, Chen E. Socioeconomicstatus and the health of youth:a multilevel, multidomain approach toconceptualizing pathways. Psychol Bull.2013;139(3):606–654

3. Williams DR, Collins C. Racial residentialsegregation: a fundamental cause ofracial disparities in health. PublicHealth Rep. 2001;116(5):404–416

4. Diez Roux AV, Merkin SS, Arnett D, et al.Neighborhood of residence andincidence of coronary heart disease.N Engl J Med. 2001;345(2):99–106

5. Sternthal MJ, Jun HJ, Earls F, Wright RJ.Community violence and urban

childhood asthma: a multilevel analysis.Eur Respir J. 2010;36(6):1400–1409

6. Suglia SF, Ryan L, Laden F, Dockery DW,Wright RJ. Violence exposure, a chronicpsychosocial stressor, and childhoodlung function. Psychosom Med. 2008;70(2):160–169

7. Midei AJ, Matthews KA. Interpersonalviolence in childhood as a risk factorfor obesity: a systematic review of theliterature and proposed pathways. ObesRev. 2011;12(5):e159–e172

8. Sundquist K, Theobald H, Yang M, Li X,Johansson SE, Sundquist J.Neighborhood violent crime andunemployment increase the risk ofcoronary heart disease: a multilevelstudy in an urban setting. Soc Sci Med.2006;62(8):2061–2071

9. Carroll JE, Gruenewald TL, Taylor SE,Janicki-Deverts D, Matthews KA,Seeman TE. Childhood abuse, parentalwarmth, and adult multisystembiological risk in the Coronary ArteryRisk Development in Young Adultsstudy. Proc Natl Acad Sci USA. 2013;110(42):17149–17153

10. Evans GW, Kim P, Ting AH, Tesher HB,Shannis D. Cumulative risk, maternalresponsiveness, and allostatic loadamong young adolescents. Dev Psychol.2007;43(2):341–351

11. Miller GE, Lachman ME, Chen E,Gruenewald TL, KarlamanglaAS, Seeman TE. Pathways to resilience:maternal nurturance asa buffer against the effects of childhoodpoverty on metabolic syndrome at

8 CHEN et al at Galter Health Sciences Library on December 19, 2019www.aappublications.org/newsDownloaded from

midlife. Psychol Sci. 2011;22(12):1591–1599

12. World Health Organization. Chronicrespiratory diseases: asthma. 2018.Available at: www.who.int/respiratory/asthma/en/. Accessed March 1, 2019

13. Wright RJ, Mitchell H, Visness CM, et al.Community violence and asthmamorbidity: the Inner-City Asthma Study.Am J Public Health. 2004;94(4):625–632

14. Wright RJ. Health effects of sociallytoxic neighborhoods: the violence andurban asthma paradigm. Clin ChestMed. 2006;27(3):413–421, v

15. Busse WW, Lemanske RF Jr. Asthma.N Engl J Med. 2001;344(5):350–362

16. Chung KF, Barnes PJ. Cytokines inasthma. Thorax. 1999;54(9):825–857

17. Spellberg B, Edwards JE Jr. Type 1/Type2 immunity in infectious diseases. ClinInfect Dis. 2001;32(1):76–102

18. Busillo JM, Cidlowski JA. The five Rs ofglucocorticoid action duringinflammation: ready, reinforce, repress,resolve, and restore. Trends EndocrinolMetab. 2013;24(3):109–119

19. Odgers CL, Caspi A, Bates CJ, SampsonRJ, Moffitt TE. Systematic socialobservation of children’sneighborhoods using Google StreetView: a reliable and cost-effectivemethod. J Child Psychol Psychiatry.2012;53(10):1009–1017

20. Sampson RJ, Raudenbush SW.Systematic social observation of publicspaces: a new look at disorder in urbanneighborhoods. Am J Sociol. 1999;105:603–651

21. Hammen C. Generation of stress in thecourse of unipolar depression.J Abnorm Psychol. 1991;100(4):555–561

22. Adrian C, Hammen C. Stress exposureand stress generation in children ofdepressed mothers. J Consult ClinPsychol. 1993;61(2):354–359

23. Miller MR, Hankinson J, Brusasco V,et al; ATS/ERS Task Force.Standardisation of spirometry. EurRespir J. 2005;26(2):319–338

24. Wang X, Dockery DW, Wypij D, Fay ME,Ferris BG Jr. Pulmonary functionbetween 6 and 18 years of age. PediatrPulmonol. 1993;15(2):75–88

25. Juniper EF, Guyatt GH, Feeny DH, FerriePJ, Griffith LE, Townsend M. Measuringquality of life in children with asthma.Qual Life Res. 1996;5(1):35–46

26. Bacharier LB, Strunk RC, Mauger D,White D, Lemanske RF Jr, Sorkness CA.Classifying asthma severity in children:mismatch between symptoms,medication use, and lung function. AmJ Respir Crit Care Med. 2004;170(4):426–432

27. McQuaid EL, Walders N, Kopel SJ, FritzGK, Klinnert MD. Pediatric asthmamanagement in the family context: thefamily asthma management systemscale. J Pediatr Psychol. 2005;30(6):492–502

28. Rosenblum Lichtenstein JH, Hsu YH,Gavin IM, et al. Environmental mold andmycotoxin exposures elicit specificcytokine and chemokine responses.PLoS One. 2015;10(5):e0126926

29. Chen E, Hanson MD, Paterson LQ, GriffinMJ, Walker HA, Miller GE.Socioeconomic status andinflammatory processes in childhoodasthma: the role of psychologicalstress. J Allergy Clin Immunol. 2006;117(5):1014–1020

30. Chen E, Fisher EB Jr, Bacharier LB,Strunk RC. Socioeconomic status,stress, and immune markers inadolescents with asthma. PsychosomMed. 2003;65(6):984–992

31. Chowdhury F, Williams A, Johnson P.Validation and comparison of twomultiplex technologies, Luminex andMesoscale Discovery, for humancytokine profiling. J Immunol Methods.2009;340(1):55–64

32. Ehrlich KB, Miller GE, Shalowitz M, et al.Secure base representations inchildren with asthma: links withsymptoms, family asthma management,and cytokine regulation [publishedonline ahead of print June 6, 2018].Child Dev. doi:10.1111/cdev.13098

33. Miller GE, Gaudin A, Zysk E, Chen E.Parental support and cytokine activityin childhood asthma: the role ofglucocorticoid sensitivity. J Allergy ClinImmunol. 2009;123(4):824–830

34. Miller GE, Chen E. Harsh family climatein early life presages the emergence ofa proinflammatory phenotype in

adolescence. Psychol Sci. 2010;21(6):848–856

35. Aiken LS, West SG. Multiple Regression:Testing and Interpreting Interactions.Thousand Oaks, CA: Sage Publications;1991

36. Duncan DT, Johnson RM, Molnar BE,Azrael D. Association betweenneighborhood safety and overweightstatus among urban adolescents. BMCPublic Health. 2009;9:289

37. Burdette HL, Wadden TA, Whitaker RC.Neighborhood safety, collective efficacy,and obesity in women with youngchildren. Obesity (Silver Spring). 2006;14(3):518–525

38. Fitzpatrick KM, Piko BF, Wright DR,LaGory M. Depressive symptomatology,exposure to violence, and the role ofsocial capital among African Americanadolescents. Am J Orthopsychiatry.2005;75(2):262–274

39. Yen IH, Yelin EH, Katz P, Eisner MD, BlancPD. Perceived neighborhood problemsand quality of life, physical functioning,and depressive symptoms amongadults with asthma. Am J Public Health.2006;96(5):873–879

40. Brody GH, Miller GE, Yu T, Beach SR,Chen E. Supportive family environmentsameliorate the link between racialdiscrimination and epigenetic aging:a replication across two longitudinalcohorts. Psychol Sci. 2016;27(4):530–541

41. Suglia SF, Enlow MB, Kullowatz A, WrightRJ. Maternal intimate partner violenceand increased asthma incidence inchildren: buffering effects of supportivecaregiving. Arch Pediatr Adolesc Med.2009;163(3):244–250

42. Chen E, Brody GH, Miller GE. Childhoodclose family relationships and health.Am Psychol. 2017;72(6):555–566

43. Liu LY, Coe CL, Swenson CA, Kelly EA, KitaH, Busse WW. School examinationsenhance airway inflammation toantigen challenge. Am J Respir CritCare Med. 2002;165(8):1062–1067

44. Marin TJ, Chen E, Munch JA, Miller GE.Double-exposure to acute stress andchronic family stress is associated withimmune changes in children withasthma. Psychosom Med. 2009;71(4):378–384

PEDIATRICS Volume 144, number 2, August 2019 9 at Galter Health Sciences Library on December 19, 2019www.aappublications.org/newsDownloaded from

45. Miller GE, Chen E. Life stress anddiminished expression of genesencoding glucocorticoid receptor andbeta2-adrenergic receptor in childrenwith asthma. Proc Natl Acad Sci USA.2006;103(14):5496–5501

46. Wright RJ, Visness CM, Calatroni A, et al.Prenatal maternal stress and cordblood innate and adaptive cytokineresponses in an inner-city cohort. AmJ Respir Crit Care Med. 2010;182(1):25–33

47. Wright RJ, Cohen S, Carey V, Weiss ST,Gold DR. Parental stress as a predictorof wheezing in infancy: a prospectivebirth-cohort study. Am J Respir CritCare Med. 2002;165(3):358–365

48. Chen E, Strunk RC, Trethewey A,Schreier HM, Maharaj N, Miller GE.Resilience in low-socioeconomic-statuschildren with asthma: adaptations tostress. J Allergy Clin Immunol. 2011;128(5):970–976

49. Chen E, McLean KC, Miller GE. Shift-and-persist strategies: associations withsocioeconomic status and theregulation of inflammation amongadolescents and their parents.Psychosom Med. 2015;77(4):371–382

50. Miller GE, Chen E, Armstrong CC, et al.Functional connectivity in centralexecutive network protects youthagainst cardiometabolic risks linkedwith neighborhood violence. Proc NatlAcad Sci. 2018;115:12063–12068

51. Miller GE, Brody GH, Yu T, Chen E. Afamily-oriented psychosocialintervention reduces inflammation inlow-SES African American youth. ProcNatl Acad Sci USA. 2014;111(31):11287–11292

52. Fisher PA, Gunnar MR, Chamberlain P,Reid JB. Preventive intervention formaltreated preschool children: impacton children’s behavior, neuroendocrineactivity, and foster parent functioning.J Am Acad Child Adolesc Psychiatry.2000;39(11):1356–1364

53. Wysocki T, Harris MA, Buckloh LM, et al.Effects of behavioral family systemstherapy for diabetes on adolescents’family relationships, treatmentadherence, and metabolic control.J Pediatr Psychol. 2006;31(9):928–938

54. Wysocki T, Harris MA, Buckloh LM, et al.Randomized trial of behavioral familysystems therapy for diabetes:maintenance of effects on diabetes

outcomes in adolescents. DiabetesCare. 2007;30(3):555–560

55. Brody GH, Kogan SM, Grange CM.Translating longitudinal, developmentalresearch with rural African Americanfamilies into prevention programs forrural African American youth. In: King R,Maholmes V, eds. The Oxford Handbookof Poverty and Child Development. NewYork, NY: Oxford University Press; 2012:553–570

56. Brody GH, Yu T, Chen E, Miller GE.Prevention moderates associationsbetween family risks and youthcatecholamine levels. Health Psychol.2014;33(11):1435–1439

57. Brody GH, Yu T, Beach SR, Philibert RA.Prevention effects ameliorate theprospective association betweennonsupportive parenting anddiminished telomere length. Prev Sci.2015;16(2):171–180

58. Chen E, Miller GE, Yu T, Brody GH.Unsupportive parenting moderates theeffects of family psychosocialintervention on metabolic syndrome inAfrican American youth. Int J Obes.2018;42(4):634–640

10 CHEN et al at Galter Health Sciences Library on December 19, 2019www.aappublications.org/newsDownloaded from

DOI: 10.1542/peds.2018-3300 originally published online July 18, 2019; 2019;144;Pediatrics 

Rachel E. Story and Gregory E. MillerEdith Chen, Robin Hayen, Van Le, Makeda K. Austin, Madeleine U. Shalowitz,

Neighborhood Social Conditions, Family Relationships, and Childhood Asthma

ServicesUpdated Information &

http://pediatrics.aappublications.org/content/144/2/e20183300including high resolution figures, can be found at:

Referenceshttp://pediatrics.aappublications.org/content/144/2/e20183300#BIBLThis article cites 55 articles, 11 of which you can access for free at:

Subspecialty Collections

http://www.aappublications.org/cgi/collection/asthma_subtopicAsthmahttp://www.aappublications.org/cgi/collection/pulmonology_subPulmonologyy_subhttp://www.aappublications.org/cgi/collection/psychiatry_psychologPsychiatry/Psychologyfollowing collection(s): This article, along with others on similar topics, appears in the

Permissions & Licensing

http://www.aappublications.org/site/misc/Permissions.xhtmlin its entirety can be found online at: Information about reproducing this article in parts (figures, tables) or

Reprintshttp://www.aappublications.org/site/misc/reprints.xhtmlInformation about ordering reprints can be found online:

at Galter Health Sciences Library on December 19, 2019www.aappublications.org/newsDownloaded from

DOI: 10.1542/peds.2018-3300 originally published online July 18, 2019; 2019;144;Pediatrics 

Rachel E. Story and Gregory E. MillerEdith Chen, Robin Hayen, Van Le, Makeda K. Austin, Madeleine U. Shalowitz,

Neighborhood Social Conditions, Family Relationships, and Childhood Asthma

http://pediatrics.aappublications.org/content/144/2/e20183300located on the World Wide Web at:

The online version of this article, along with updated information and services, is

http://pediatrics.aappublications.org/content/suppl/2019/07/17/peds.2018-3300.DCSupplementalData Supplement at:

ISSN: 1073-0397. 60007. Copyright © 2019 by the American Academy of Pediatrics. All rights reserved. Print the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois,has been published continuously since 1948. Pediatrics is owned, published, and trademarked by Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it

at Galter Health Sciences Library on December 19, 2019www.aappublications.org/newsDownloaded from