Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Sep 3.
Published in final edited form as: Soc Sci Med. 2019 Mar 3;271:112203. doi: 10.1016/j.socscimed.2019.02.054

Associations between neighborhood and family factors on symptom change in childhood attention deficit hyperactivity disorder

Wendy Sharp 1, Aman Mangalmurti 2, Carlisha Hall 2, Saadia Choudhury 2, Philip Shaw 1,2
PMCID: PMC6748875  NIHMSID: NIHMS1042312  PMID: 30857751

Abstract

Transactional theories view development as partly shaped by processes proximal to a child, which in turn interact with more distal neighborhood and societal contexts. Here we apply this theory to parse the interplay between neighborhood and familial factors on age-related change in symptoms of inattention and hyperactivity-impulsivity (ADHD). A cohort of 190 children (96 with ADHD) had a range of neighborhood and familial factors ascertained and had repeated clinical assessments over an average of 2.5 years at a U.S. research center. Using mixed model regression, we found an association between neighborhood wealth, but not the built environment, and the annual rate of change of inattentive but not hyperactive-impulsive symptoms. Following the transactional model, we asked if familial processes explain (mediate), modify (moderate), or act alongside this effect of neighborhood wealth on the change in a child’s symptoms of inattention with age. We found evidence for moderation. Specifically, several family level variables – parental economic/education status and degree of family conflict and order moderated the effects of neighborhood wealth on the change in a child’s inattentive symptoms. Children living in relatively affluent neighborhoods showed improvement with age in inattention, largely independent of variation in a wide range of familial factors. By contrast, children living in less affluent neighborhood showed clinical deterioration only if the family had high levels of conflict or if the parents were of lower economic/educational status. Such work might help identify children whose familial and neighborhood contexts place them at risk of having ADHD symptoms persist or increase with age.

Keywords: ADHD, transactional model, neighborhood wealth, built environment, family conflict, SES, symptom trajectories

Introduction.

Attention deficit hyperactivity disorder (ADHD) affects between 4–10% of children worldwide, constituting a major public health challenge (Polanczyk et al., 2007). Around 20% of these children will continue to have the full syndrome as adults, a further 50% will continue to show impairing symptoms and the remainder show full remission by adulthood (Faraone et al., 2006). Understanding the factors that contribute to this variable clinical course could help identify points for intervention, not only within the affected child, but the family and the broader social context.

Genetic factors are important in the onset of ADHD. Family, twin, and adoption studies show that childhood ADHD has a heritability rate of around 70% and common genetic variants have been recently identified that confer risk at a genome wide level of significance (Demontis et al., 2017; Faraone & Larsson, 2018). Genetic factors also largely determine the change in ADHD symptoms that occurs with age. A large longitudinal twin study found that 81% of the change in hyperactivity-impulsivity and 50% of the change in of inattention between ages 8 and 16 is determined by genetic factors (Pingault et al., 2015). There were modest but significant contributions from non-shared and shared environmental factors which include neighborhood and familial factors.

It is important to consider these familial and neighborhood contexts as the interplay between these environments and genetic factors impacts on symptom expression, partly through epigenetic modifications that can alter gene expression (Walton et al., 2017).

We frame our examination of how neighborhood and families might impact upon the change with age in a child’s symptoms of ADHD within the framework proposed by Bronfenbrenner (Bronfenbrenner, 2009). In this model, a child’s development is seen as driven by proximal processes, such as the family, that unfold over time. The power of these proximal processes varies as a function of properties of the child (including the child’s genotype, sex and age) and more distal neighborhood and societal contexts. In this study we examine how age-related change in ADHD symptoms is influenced by the interplay of neighborhood and family variables (Johnston & Chronis-Tuscano, 2015). First, we review the literature to identify the aspects of the neighborhood that have been most robustly associated with ADHD and related behavioral disorders. We then outline three models of how neighborhood and familial factors might impact the symptoms of ADHD over time. The first model holds that neighborhood and familial factors have independent, additive effects on the change with age in ADHD symptoms. The second model posits that familial factors explain or mediate the association between neighborhood and a child’s symptom trajectories. The final model holds that familial factors proximal to the child modify or moderate the effects of more distal neighborhood contexts.

Neighborhoods and ADHD.

Many studies have examined associations between neighborhood level variables and ADHD as well as other externalizing disorders such as conduct disorder. For instance, a Swedish, nationwide birth cohort study found that 5% of the variance in childhood behavioral outcomes could be attributed to neighborhood factors after controlling for family/individual level variables (Sundquist et al., 2015). Additionally, the study showed that effects of neighborhood deprivation increased the risk of childhood ADHD by a third, though this effect was smaller than for conduct disorder in which risk was doubled. A separate study using data from the 2011 U.S. National Survey of Children’s Health found a linear increase in risk of a diagnosis of childhood ADHD with increasing severity of neighborhood adversity (Dahal et al., 2018). While both of these studies showed that neighborhood factors explained less variance in childhood behavioral problems than family-level factors, nonetheless the neighborhood effects were significant. It is thus meaningful to examine how they might act on children with ADHD.

Several studies have asked the question of which neighborhood factors are important to the development and clinical course of behavioral disorders. Neighborhood economic deprivation has been identified as a factor that increases the odds of ADHD and related behavioral disorders (Kohen et al., 2008; Leventhal & Brooks-Gunn, 2000; Razani et al., 2015; Xue et al., 2005). The impact of neighborhood wealth remains after controlling for the wealth of families. Indeed, the mismatch between neighborhood and family affluence in itself is associated with adverse mental health outcomes (Vilhjalmsdottir et al., 2016). In addition to measures of neighborhood wealth, the nature of the built environment can impact on a child’s mental health. Some studies find that children living in neighborhoods characterized by vandalism and dilapidated housing are at greater risk for ADHD and disruptive behaviors. For example, using data on 65,000 children from the National Survey of Children’s Health, children living in neighborhood with poor physical characteristics showed a 44% increased risk of ADHD or other disruptive behavioral disorders (Butler et al., 2012). Additionally, fewer ADHD symptoms and better performance on attention demanding tasks are associated with more ready access to, and more time spent playing in green spaces (Amoly et al., 2014; Taylor & Kuo, 2009). Finally, cumulative exposure to road traffic noise during childhood has been tied to more ADHD symptoms (Hjortebjerg et al., 2016). In sum, these studies delineate preliminary cross-sectional associations between different individual environmental exposures and ADHD. We advance the literature by considering the effects of a wide range of neighborhood factors, specifically neighborhood wealth, poverty rates, crime rates, access to green spaces and areas for physical activity, school enrollment and housing prices on age-related change in ADHD symptoms.

The interplay of neighborhood and families.

How do neighborhood factors impact the child? The first possibility is that neighborhood and family level factors exert independent, additive effects on a child’s clinical course. Such additive main effects have been reported. For example, using data from a community survey of 80 neighborhoods, Xue and colleagues found additive main effects of maternal characteristics and neighborhood disadvantage (indexed by poverty rate and residential instability) on a child’s internalizing problems (Xue et al., 2005) Here we determine if similar additive neighborhood and familial effects influence the change in ADHD symptoms with age.

Secondly, we ask if family level processes mediate the relationship between neighborhood contexts and ADHD outcome. Mediation models are consistent with transactional models as the child is nested within families which in turn are nested within neighborhoods. A natural extension of this ‘concentric’ organization is to hypothesize that neighborhood contexts are ‘translated’ into the family context and then to the child. In an influential review, mediation of neighborhood by familial factors was posited by Leventhal and Brooks-Gunn as one of the three principle ways in which neighborhoods impact on children (Leventhal & Brooks-Gunn, 2000). They argue this mediation route is complemented by more direct neighborhood impacts on children, such as the availability of community resources and the impact of community norms (e.g. the level of formal and informal monitoring of a child by community members). Mediation models are also particularly apt when independent variables (here, relatively static neighborhood factors) at least partly precede the dependent variable (here, the change in ADHD symptoms) (MacKinnon et al., 2007). Such temporal precedence is compatible with a directional link from the independent to dependent variables. Many theorists have also considered the exact mechanisms through which familial factors could mediate neighborhood disadvantage. Many adverse neighborhood features can act as stressors for the family, such as a family’s exposure to violence, housing instability and job uncertainty (Lipina & Colombo, 2009). The familial response to these stressors could impact on a child’s behavioral outcome (McLoyd, 1998; Phillips & Shonkoff, 2000).

Evidence for mediation comes from a longitudinal twin study that found the association between neighborhood socioeconomic status and children’s antisocial behavior to be completely mediated by maternal warmth and parental monitoring (Odgers et al., 2012). There are other examples of how familial characteristics (such as parental distress, family conflict and parenting style) mediate the association between neighborhood adversity (both perceived and objective) and a child’s externalizing behavior problems, cross-sectionally and over time (Doan et al., 2012; Li et al., 2017; Plybon & Kliewer, 2001). These studies, while informative, considered externalizing behavior problems, assessed by parental questionnaires, rather than assessing clinical status using the gold standard of a clinician administered structured interview. Here we further the literature through our consideration of a wide range of measured neighborhood factors using detailed clinical assessments of ADHD collected longitudinally.

A third model holds that that familial factors modify, or moderate, the impact of neighborhood factors. Concepts of moderation underpin several leading theories of how neighborhood and families interact to affect a child’s mental health. For example, several theorists have argued that advantageous familial characteristics can protect or buffer the child from the impact of adverse neighborhood factors (Chase-Lansdale & Gordon, 1996; McKelvey et al., 2011; Silk et al., 2004). Other theorists have emphasized the potentiating (i.e. multiplicative rather than additive) effects of neighborhood and familial disadvantage (Garbarino & Sherman, 1980).

The moderation model has strong empirical foundations. Twin studies have found that the degree of parental involvement moderates the influence of environmental factors on the child’s ADHD symptoms, as reported by the mother (Nikolas et al., 2012). When parental involvement was high there was little impact of the non-shared environmental contributions, which could include neighborhood levels factors uniquely affecting each child, on ADHD. As parental involvement decreased, the impact of non-shared environment increased. Such twin studies do not specify the exact environmental variables but point to their importance en masse. A delineation of candidate family factors comes from a prospective epidemiological study of 16,000 children in the UK (Flouri et al., 2015). The study found that the effects of neighborhood disadvantage (indexed by the proportion of families in poverty, crime and air quality) on childhood behavioral problems were moderated by parenting style. Specifically, parenting characterized by warmth and sensitivity buffered the adverse effects of neighborhood on the child over time. While there are other demonstrations of interactions between neighborhood and family level factors (Burchinal et al., 2006; Dearing, 2004), none have examined ADHD as a specific outcome, nor how these familial factors may be associated with the course of the disorder.

Family level factors

Next, we ask which familial factors are most likely to act as moderators, mediators or main effects. Several family level features have emerged as being of particular interest. First, a recent meta-analysis found that children in families of low socio-economic status were approximately twice as likely to have ADHD as children from high income families (Russell et al., 2016). This increased risk held after controlling for comorbidities such as conduct disorder (Miller et al., 2017). Secondly, several studies have examined the impact of parental ADHD on children (Johnston & Chronis-Tuscano, 2015; Johnston & Mash, 2001). Around 60% of affected parents will have an affected child and around 40% of children with ADHD have an affected parent (Minde et al., 2003). The increased risk of ADHD in offspring of parents with ADHD can in large part be attributed to genetics, given the high heritability of ADHD. However, studies of biologically unrelated mothers and children (from donor conception or adoption at birth studies) have also reported associations between a mother’s and a child’s symptoms of ADHD, largely mediated by parenting style (Harold et al., 2013). This finding is consistent with prior observations of how parental ADHD impacts parenting style. One meta-analysis suggests that parents with ADHD exhibit parenting styles that are either excessively ‘harsh’, (high on familial conflict and control) or ‘lax’ (low on both control and involvement) (Park et al., 2017). Harsh parenting styles have been further associated with the severity of a child’s ADHD (for a review see (Modesto-Lowe et al., 2008). In summary, there is robust evidence for associations between a child’s ADHD and both parental socio-economic status and parental ADHD.

We also consider measures of family environment derived from the Family Environment Scale (Moos, 1990). This parent-completed questionnaire assesses ten aspects of family life: familial cohesion, organization, conflict, control, participation in recreational activity, interest in cultural/intellectual activities, focus on achievement, emphasis on ethical and religious issues, familial self-sufficiency and familial encouragement to express feelings. We used this scale given prior reports that families of children with ADHD show higher levels of family conflict and lower familial cohesion (as measured by the scale) which held after controlling for familial socioeconomic status and parental psychopathology (Biederman et al., 1995). Others have found that family conflict both accounted for up to 40% of the impairment seen in children with ADHD (Pressman et al., 2006) and exacerbated the comorbidities seen in the child with ADHD (Kepley & Ostrander, 2007).

Individual level effects: child age and gender.

Two child (individual) level factors are important to consider. Firstly, it could be argued that neighborhoods will have less impact on the mental health of very young children as they are insulated from these effects by parents. Evidence on this point is mixed. Some studies find a minimal impact of neighborhood contexts on behavioral outcomes until the end of elementary school, around 11 years of age (Humphrey & Root, 2017). However, other studies find significant neighborhood associations with childhood behavioral outcomes in children as young as 6 (Chase-Lansdale & Gordon, 1996). Thus, we consider age in our analyses by asking if it interacts with either family or neighborhood level factors in determining age-related change in ADHD symptoms.

Gender is also important to consider as boys and girls might have different exposures to risk factors pertinent to ADHD at both the neighborhood and family levels. Indeed, some studies report that boys are exposed to more community violence and harsher parenting practices(Farrell & Bruce, 1997; Lytton & Romney, 1991; Selner-O’Hagan et al., 1998); however, some others do not (Margolin & Gordis, 2000; Moffitt et al., 2001). Additionally, one meta-analysis found that boys and girls do not differ significantly in the association between familial conflict and adverse behavioral outcomes (Kitzmann et al., 2003). Here, we ask if gender interacts with either family or neighborhood level factors in determining the change in ADHD symptoms.

Age-related change in symptoms as an outcome measure

Our primary outcome measure is the age-related change in symptoms of inattention and hyperactivity-impulsivity, specifically the annual rate of change in number of symptoms. While psychiatric practice views the age-related changes in ADHD in categorical terms (persisters versus remitted) (A.P.A., 2013) we consider the change in symptom dimensions for three reasons. First, epidemiological evidence suggests that the symptoms of ADHD can be considered dimensionally, lying at the extreme end of a continuous distribution of symptoms (Lubke et al., 2009; Polderman et al., 2007). Secondly, assessing age-related change in a quantitative trait, such as the number of symptoms, provides a more precise measure of clinical course than is obtained through considering changes in dichotomous diagnostic status alone. Thirdly, inattention and hyperactivity-impulsivity are considered separately in view of their partly distinct genetic etiologies, cognitive substrates, comorbidities and dissociable developmental trajectories (Kuntsi et al., 2014; Larsson et al., 2011; Weiler et al., 2000; Willcutt et al., 2007; Wood et al., 2009).

In summary, adopting a transactional framework, we aim to parse the interplay between neighborhood factors and familial processes in determining the rate of change of child’s ADHD symptoms. We ask if familial processes explain (mediate) or modify (moderate) the effects of neighborhood contexts on a child’s ADHD, or if neighborhood and familial factors have independent additive effects.

Methods

Cohort.

The study cohort comprised 190 children from 154 nuclear families: there were 26 sibling pairs and 5 sets of three siblings. The children (128 boys [67%] and 62 girls [33%]) entered the study at a mean age of 8.5 years (median 8.2, S.D. 0.9) and were followed for an average 2.6 years (SD 1.4). Thus, the mean age at final assessment was 11.1 years (median 10.9, SD. 2.7 ). Self-ascribed race was obtained from parents: 129 (68%) identified as white, 19 (10%) as African American; 21 (11%) as mixed-race, 7 (4%) as Asian and one as Native American. Eighteen (10%) identified their ethnicity as Hispanic. The general inclusion criteria were (1) between ages 3 and 18 at study entry (2) Intelligence Quotient of 70 or greater, defined using age appropriate version of the Wechsler intelligence scales. General exclusion criteria were (1) neurological disorders known to affect movement and attention (except for tic disorder); (2) psychiatric disorders other than ADHD, oppositional defiant disorder, conduct disorder, major depression or anxiety disorders.

We ascertained the symptoms of inattention (minimum of zero and maximum of nine) and hyperactivity-impulsivity (minimum of zero and maximum of nine) using the Diagnostic Interview for Children and Adolescents, a clinician administered interview with parents (Reich, 2000). Six or more symptoms in either domain warrants a diagnosis. All interviews were conducted by experienced clinicians (PS is a psychiatrist and WS is a clinical social worker; interrater reliability of K > 0.9). All children were clinically assessed at the Clinical Center of the NIH at least twice; the ages for each assessment are illustrated graphically in Figure 1.

Figure 1.

Figure 1

Figure 1

Open in a new tab

A and B. Among those with ADHD, the annual rate of change of inattention was highly variable: 45 (47%) showed improvement, 42 (44%) had worsening symptoms and 9 (9%) showed no age-related change. Among those who did not meet criteria for ADHD group, 63 (67%) entered the study with no symptoms of inattention and remained symptom free. Twenty-one (22%) showed the development or worsening or symptoms of inattention during follow-up (but never met criteria for ADHD) and 10 (11%) showed improvement from baseline symptoms.

C and D Symptoms of hyperactivity-impulsivity showed more marked age related decline, with an overall mean decrease during this childhood age window of 0.29 (SD 0.68). When confined to the 96 subjects who had ADHD, the mean rate of annual decrease in hyperactivity-impulsivity was 0.54 (SD 0.73). Sixty nine of those with ADHD (72%) showed improvement in hyperactivity-impulsivity, 15 (16%) showed no change and 12 (12%) showed worsening symptoms. For those who did not meet criteria for ADHD, 76 (81%) remained free from hyperactivity-impulsivity throughout, while 7 (7%) developed symptoms during follow-up and 11 (12%) showed improvement of symptoms during follow-up.

The procedures were approved by the Institutional Review Board of the National Human Genome Research Institute. Parents or legal guardians of children gave informed, written consent for children who participated (under 18 years of age). The child gave verbal assent and written assent (very young children could provide just the first letter of their first name etc.). Participants who reached 18 years of age during the study provided informed, written consent.

Assessment of neighborhood and family factors.

Ten neighborhood level variables were extracted from publicly available databases based on the family’s address. Income (median household income in US dollars in the past 12 months), poverty rates (percentage of population below poverty level in past 12 months), and school enrollment (percentage of children over age 3 enrolled in public or private school) metrics were determined by zip code and based on 2015 U.S. Census data. (factfinder.census.gov). Estimates for home value were determined from real-estate marketplace estimates. Crime rates were determined using address specific results (from AreaVibes: areavibes.com) which retrieved data from the FBI’s Uniform Crime Reports for 2016. Park access (percent of people living within half a mile of a park) and air quality (particulate matter in μg/m3) measurements are county estimates taken from the CDC’s National Environmental Public Health Tracking Network (ephtracking.cdc.gov/). Obesity rates (percentage of adults with a BMI of 30 or more), a food environment index (indicating limited access to healthy foods and food insecurity) and an exercise access metric (percentage of population with adequate access to locations for physical activity) were pulled from the County Health Rankings & Roadmaps program (countyhealthrankings.org).

Principal component analysis (using a varimax rotation, retaining components with eigenvalues greater than one) reduced these variables to three independent components, as shown in Table 1. The components reflected (1) ‘neighborhood wealth’ (highest loadings on proportion of families in poverty, median family income, food environment index, crime rates); (2) ‘access to health promoting activities’ (loadings from access to green spaces, areas for physical activity, obesity rates); (3) ‘schools and housing’ (loadings from local housing costs and proportion of children enrolled in state schools).

Table 1.

Principal component analysis reduced (a) 10 neighborhood variables to three independent components; (b) 13 family level variables to 4 components. Variables loadings above 0.3 onto each component are shown.

  (A) Principal component analysis for neighborhood variables
Factor 1:
Neighborhood
wealth		 Factor 2: Access
to health
promoting
activities		 Factor 3:
Housing/school
Variance Explained 26.7% 25.9% 19.8%
Crime rate 0.87
Proportion of families in poverty 0.85
Family income 0.68 0.53
Food index 0.56 0.47
Access to green spaces 0.90
Access to adequate locations for physical activity 0.89
Proportion of obese individuals 0.35 0.82
Proportion of children in public school 0.85
Home price 0.38 0.69
Air quality 0.31 0.52
  (B) Principal component analysis for family level variables
Factor 1:
Familial control
/cohesion		 Factor 2: Familial
self-sufficiency
and ADHD history		 Factor 3: Family
ethics and
economic
status		 Factor 4:
Familial
conflict
Variance explained 15.3% 14.6% 12.5% 9.5%
Familial control 0.65
Familial organization 0.60
Familial achievement 0.59
Familial cohesion 0.57
Familial expressivity 0.40 0.38 0.36
Wenders Utah Rating Scale 0.85
Parental ADHD History 0.85
Familial independence 0.41 0.32
Parental economic/educational status 0.71
Familial ethics 0.63
Familial activity 0.42 0.42
Familial conflict 0.72
Intellectual/cultural pursuits 0.46 0.70
Open in a new tab

Family level factors.

Parental economic/educational status was calculated using the Hollingshead scale which assesses parental educational attainment and occupation (Hollingshead, 1975). Parental history of ADHD was based on report of a diagnosis of ADHD made by a physician/clinical psychologist. This was supplemented by the Wenders Utah rating scales, a standardized questionnaire given to adults concerning their childhood symptoms of ADHD (Ward et al., 1993). Parental economic/educational status and parental ADHD were examined in separate models, given their theoretical importance, as discussed earlier.

Ten aspects of the family environment were further assessed using the Family Environment Scale, completed at study entry by a parent (Moos, 1990). We combined these 10 measures with the three measures of parental economic/educational status and parental ADHD and used PCA to reduce data into four components (using a varimax rotation, retaining factors with eigenvalues greater than one) – Table 1b. The components reflected (1) the degree of familial order (loadings from items of family control, cohesion and organization); (2) items pertaining to the family self-sufficiency and parental ADHD; (3) parental economic/education status along with the family’s emphasis on ethical issues; and (4) familial conflict. A child’s medication history was obtained from parents. ADHD medication use was expressed as the proportion of the total time during the study that the child was taking regular psychostimulant medications.

Outcome measure: annual rate of change in ADHD symptoms.

The primary outcome variables were a child’s annual rate of change in symptoms of inattention and of hyperactivity-impulsivity. Rates were calculated using ordinary least squares regression and we considered only linear models as 107 subjects (56%) had only two observations. We thus did not consider non-linear trajectories at the individual level as such analyses require at least three and ideally four or more observations on all participants. Thus change in symptoms was represented by one value: a rate of change. A negative value indicates decreasing symptoms and thus a better clinical course; zero indicates no change in symptom severity; a positive value indicates age- related increase in symptoms.

Analysis.

We used mixed model regression to test for associations between predictor variables and outcomes, including a random term for family identity (to account for the non-independence of observations). The lmer R package was used (Bates et al., 2014). To test for effects of child age and gender, these were allowed to interact with the neighborhood and family factors in the determination of the outcome variable (annual rate of symptom change).

We next tested the three models: namely whether (1) family and neighborhoods factors had additive main effects; (2) family level factors mediated the associations between neighborhoods and clinical outcomes; (3) or if family levels factors moderated this association. To test for additive main effects, the family and neighborhood levels factors were entered jointly into the mixed model regression. To test for moderation, interactions terms between the neighborhood and family level variables were included in the regression models. To test for mediation, a series of mixed model regressions parsed the associations between neighborhood factors, outcomes and putative mediating family variables. Analyses used the R package Mediation (Imai et al., 2010). In the mediation analysis, a bootstrapping approach with 1,000,000 resamples was utilized to provide a bias corrected confidence interval for estimates. Mediation effects were considered significant if 95% confidence intervals did not cross zero. We tested for the effect of parental economic/educational status, parental ADHD, and also the four composite family level variables.

To test for developmental effects, analyses allowed the child’s age at study entry to interact with the neighborhood and family factors in the determination of outcome. The same approach was used to test for gender effects. We also repeated analyses retaining only the youngest member of each family to ensure independence of data in a linear regression model. Similarly, we repeated analyses retaining one family from each zip code. We included race/ethnicity in regression analyses as a main effect (recoding children as belonging to the largest subpopulation; or to the other subpopulations combined). We considered race and ethnicity as a main effect only, as all subpopulations except for the white, non-Hispanic group were too small to allow for a consideration of mediating and moderating effects.

Results

During the study, there was a mean overall decrease of 0.01 (SD 1.03) symptoms of inattention per year for the entire cohort, with a decrease of 0.13 (SD 1.29) symptom/year for those meeting diagnostic criteria for ADHD and an increase of 0.11 (SD 0.66) symptom/year for those who did not. As can be seen from Figure 1, among the 96 with a diagnosis of ADHD, inattention showed a highly variable trajectory, with 45 (47%) showing improvement, 42 (44%) had worsening symptoms and 9 (9%) showed no age-related change. Among the 94 who never met criteria for ADHD, 63 (67%) entered the study with no symptoms of inattention and remained symptom free. Twenty-one (22%) showed worsening symptoms of inattention during follow-up and 10 (11%) showed improvement from baseline symptoms.

Symptoms of hyperactivity-impulsivity showed more uniform age-related decline, with an overall mean decrease of 0.29 (SD 0.68) symptoms/year. Sixty-nine of those with ADHD (72%) showed improvement in hyperactive-impulsive symptoms, 15 (16%) showed no change and 12 (12%) had worsening symptoms. This translated to a mean decrease for those with ADHD of 0.54 (SD 0.73) symptoms/year (i.e. an approximate decrease of one symptom every 2 years). For those who did not meet criteria for ADHD, 76 (81%) remained free from hyperactivity-impulsivity throughout, while 7 (7%) developed symptoms during follow-up and 11 (12%) showed improvement of symptoms during follow-up.

The rate of change of inattention and change in hyperactivity-impulsivity were modestly correlated (r=0.22, p=0.002). Eight children who entered the study without a diagnosis showed symptom exacerbation and met criteria for ADHD during follow-up (meeting diagnostic criteria between 7.9 to 12.8 years of age). Ninety-four children (49%) were never diagnosed with ADHD.

In summary, during childhood, symptoms of hyperactivity-impulsivity mostly improved whereas inattention followed a more variable clinical course.

Main effect of neighborhood factors on rate of change of symptoms.

Mixed model regression was used to determine if neighborhood wealth, access to health promoting resources and housing/schools were associated with symptom change. For change in inattention, more rapid symptomatic improvement was associated with greater neighborhood wealth (β=−0.25 SE =0.07, t=3.35, p=0.002) but not with access to health promoting resources (β= −0.05 SE =0.07, t=0.7, p=0.49) or housing/schools (β=0.11, SE=0.08, t=1.4, p=0.17). No significant associations emerged between the rate of change of hyperactivity-impulsivity and any of the neighborhood variables. There were no significant interactions between age at study entry and the neighborhood variables in association with the rate of symptom change (all p>0.05). There were also no interactions between gender and neighborhood variables (all p>0.05).

Family level processes and symptom change.

Rate of change in inattention, but not hyperactivity-impulsivity, was significantly associated with parental ADHD, with greater improvement seen among children who had one or both parents with ADHD (β=−0.42 SE =0.2, t=2.08, p=0.04). There was no association between parental economic/educational status and either the rate of change of inattention (β=−0.12, SE 0.08, t=1.52, p=0.14) or hyperactivity-impulsivity (β=−0.18, SE 0.13, t=1.38, p=0.18). Next the composite family level variables were examined. Associations emerged only for change in inattention with the degree of familial conflict (β=0.3, SE 0.07, t=4.02, p=0.0003) and familial self-sufficiency/parental ADHD (β=−0.18, SE 0.07, t=2.47, p=0.02). Neither age at study entry nor gender interacted significantly with family level variables in determining the rate of change of symptoms (all p>0.05).

The interplay of family and neighborhood level factors.

Next it was determined if family-level variables acted as mediators or moderators in the relationship between neighborhood wealth and change in inattention, or if social and family context had purely additive effects.

Additive main effects.

We tested for additive main effects by entering in one model all the family and neighborhood variables. In this model, rate of change in inattention was significantly associated only with neighborhood wealth (β=−0.25 SE=0.08, p=0.002) and familial conflict (β=0.32 SE=0.07, p=0.0001). Analyses were repeated including race/ethnicity as a main effect along with child-level variables (sex, age at study entry and symptom severity at baseline). The pattern of results held – see Supplementary Table 1. No associations were found between rate of change of hyperactivity-impulsivity and neighborhood or family factors in this model.

Mediation models.

We asked if any family level factor mediated the significant association between neighborhood wealth and rate of change of inattentive symptoms. Neither parental economic/educational status nor parental ADHD emerged as significant mediators (the 95% confidence intervals for the mediating effects all crossed zero). Similarly, none of the four composite measures of family factors acted as mediators - see Supplementary Table 2.

Moderation.

Next, family variables were considered as moderators of the association between neighborhood wealth and outcome – see Table 2. Parental family economic/educational status, emerged as a significant moderator (β=0.12, SE=0.04, t=2.52, p=0.02). This interaction, shown in Figure 2, indicates that children living in neighborhoods that were relatively affluent showed clinical improvement, regardless of familial economic/educational status. By contrast, the clinical course in children from areas of relative neighborhood disadvantage varied by parental economic/educational status, with worsening inattention seen mainly among children living in families and neighborhoods that were less affluent. The presence of parental ADHD did not act as a moderator.

Table 2:

Moderation models. Each family level factor is tested as a potential moderator of the significant association between neighborhood wealth and change in symptoms of inattention.

Model: parental education/economic status as a moderator
Inattention
Fixed effects β Std. Error P-Value
Neighborhood wealth −0.096 0.10 0.35
Parental economic/educational status 0.0038 0.082 0.96
Neighborhood wealth * Parental educational/economic status 0.12 0.049 0.02
Model: parental ADHD as a moderator
Inattention
Fixed effects β Std. Error P-Value
Neighborhood wealth −0.26 0.080 0.003
Parental ADHD −0.42 0.19 0.04
Neighborhood wealth * Parental ADHD −0.022 0.27 0.94
Model: familial factors (from PCA) moderators
Inattention
Fixed effects β Std. Error P-Value
Neighborhood wealth   −0.16 0.093  0.09
Familial control/cohesion   0.080 0.071 0.27
Familial self-sufficiency and ADHD History   −0.15 0.073 0.05
Family ethics and economic status   0.014 0.079  0.86
Familial conflict   0.34 0.076  0.0001
Neighborhood wealth * Familial control/cohesion −0.19 0.088  0.04
Neighborhood wealth * Familial self-sufficiency /ADHD History −0.046 0.088 0.61
Neighborhood wealth * Familial ethics and economic status 0.066 0.061 0.29
Neighborhood wealth * Familial conflict −0.16 0.065 0.02
Model: familial factors (from PCA) as moderator, adjusting for race/ethnicity and child level factors
Inattention
Fixed effects β Std. Error P-Value
Neighborhood wealth −0.10 0.091 0.28
Familial control/cohesion 0.036 0.069 0.60
Familial self-sufficiency and ADHD History −0.094 0.074 0.22
Family ethics and economic status 0.023 0.079 0.77
Familial conflict 0.30 0.073 0.0004
Symptoms at baseline −0.078 0.022 0.002
Sex 0.066 0.15 0.66
Age at baseline −0.061 0.027 0.04
Race/Ethnicity 0.23 0.16 0.14
Neighborhood wealth * Familial control/cohesion −0.19 0.085 0.04
Neighborhood wealth * Familial self-sufficiency /ADHD History 0.013 0.086 0.88
Neighborhood wealth * Familial ethics and economic status 0.035 0.060 0.56
Neighborhood wealth * Familial conflict −0.14 0.063 0.04
Open in a new tab

Figure 2:

Figure 2:

Open in a new tab

Moderation by familial factors of the association between neighborhood wealth and the trajectories of inattention. In each figure the lines represent the upper and lower fifth centile of each family variable [lines derived from parameters given in Table 1]. Positive values of change in inattention translate to a deterioration in clinical disorder; negative values represent an improvement. Significant moderation was found for (A) parental economic/educational status; (B) familial conflict; (C) familial control/cohesion.

Next the potential moderating effects of the four family environmental variables were examined. A significant interaction emerged between neighborhood wealth and the component reflecting familial conflict with respect to inattentive symptom change (β−0.16 SE=0.07, t=2.5 p=0.02). Children in affluent neighborhoods showed overall clinical improvement, at all levels of familial conflict. Children in relatively less affluent neighborhoods showed worsening inattention only when the family had high levels of conflict and were less engaged in cultural activities. An interaction between neighborhood wealth and the component reflecting familial order also emerged for change in inattention (β−0.19 SE=0.08, t=2.2 p=0.04). Children from areas of relative neighborhood wealth showed clinical improvement, at all level of familial order. By contrast, children from relatively disadvantaged neighborhoods showed clinical deterioration only when the family showed high degrees of familial control/cohesion.

Robustness analyses.

We repeated analyses retaining the youngest child for each family, and the pattern of results held -Supplemental Table 3a. Similarly, the overall pattern of results held when only one family from each zip code was retained -Supplemental Table 3b. ADHD medication use was expressed as the proportion of the total time during the study that the child was taking regular psychostimulant medications. This medication variable was not significantly associated with either change in inattention (β=−0.43, SE=0.43, t=1.0 p=0.34) or change in hyperactivity-impulsivity (β=0.15, SE=0.24, t=0.64, p=0.54). Twenty-two children were treated with other psychotropic medications during the study (atomoxetine 3, guanfacine 17, selective serotonin reuptake inhibitors 6, lithium 2; some children took more than one medication). We repeated analyses excluding these children and the overall pattern of results did not change – Supplementary Table 4.

Discussion

To our knowledge, this is the first demonstration of how neighborhood processes distal to a child can interact with more proximal, familial factors to influence age-related change in symptoms of ADHD. Here we find an association between neighborhood wealth and change in a child’s inattention. The change in symptoms was not mediated by family level factors, but rather by the familial variables of parental economic/education status and degree of family conflict and control. Specifically, children living in affluent neighborhoods showed overall clinical improvement at all levels of these family factors. By contrast, children in less affluent neighborhoods showed clinical deterioration only if the parents were also economically disadvantaged, or the family showed high levels of conflict or control/cohesion. Thus, an association between worsening inattention and lower neighborhood affluence was buffered by lower levels of familial conflict and less familial control/cohesion.

Neighborhood contexts and the course of inattentive symptoms.

While associations between neighborhood environments and risk for ADHD onset have been reported elsewhere (Butler et al., 2012; Razani et al., 2015), we further show an impact of neighborhoods on age-related changes of ADHD symptoms. We find neighborhood wealth is associated with the course of inattention, not hyperactivity-impulsivity, and this association held when family and child level factors were controlled. Such specificity is perhaps surprising as most studies find neighborhood contexts to be associated with a broad range of adverse childhood mental health outcomes (Conway et al., 2018). The specific associations we report might reflect the depth of the clinical phenotyping in our study, which used clinician interviews at several time points, whereas studies reporting broad effects of neighborhood contexts more often used screening questionnaires at one time point only.

While neighborhood wealth was associated with ADHD symptom course, factors reflecting the built environment and access to health promoting activities were not. Our assessment of the built environment was limited to publicly available data and we may have missed associations with unmeasured environmental variables such as specific air pollutants or neighborhood dilapidation (Amoly et al., 2014). Others have also emphasized the importance of the subjective perception of neighborhood qualities, such as perceived safety and cohesion in addition to the objective metrics we considered (Derauf et al., 2015; Rocha et al., 2017). While we did not find access to health promoting activities to improve ADHD symptom course, this does not belie the myriad other health benefits of promoting a healthy lifestyle for children (Mura et al., 2015). In summary, we expand the literature on associations between neighborhood wealth and mental health to include an impact on the course of a child’s symptoms of ADHD.

Transactional models applied to the developmental course of symptoms of ADHD.

Our primary aim was to delineate how familial factors proximal to the child might either explain or modify more distal neighborhood contexts. We find a moderation of the association between neighborhood or ‘distal’ wealth on the child’s symptom course by family or ‘proximal’ wealth. Similar moderation effects have been reported in other mental health contexts (Buu et al., 2009; Kemp et al., 2016). Our study further demonstrates that impacts of neighborhood and familial wealth can occur across a wide range of economic conditions and are not just confined to extremes of poverty.

We also find that neighborhood disadvantage was associated with symptom deterioration only among children from families that showed high levels of conflict. Such intra-familial conflict could involve expressions of hostility, which is a core aspect of ‘high expressed emotion’, a construct that in turn is tied to the exacerbation of a wide range of mental disorders in young adults (Peris & Baker, 2000). By contrast, familial concord appeared to be a protective factor for the child growing up in a neighborhood with relative economic disadvantage. There was also an interaction between neighborhood wealth and measures of familial order: the worst clinical course was associated with a combination of neighborhood disadvantage and a high degree of familial order. This is unexpected as clear family rules, routines and expectations are often seen as beneficial for a child with ADHD (Ellis & Nigg, 2009; Murray & Johnston, 2006). However, it is also possible that family order can merge into a maladaptive rigidity, with parents being less responsive to the needs of their inattentive child.

In summary, neighborhood economic advantage was associated with favorable symptom trajectories in children and this held regardless of family level factors more proximal to the child. The deleterious impact of neighborhood disadvantage was attenuated or buffered by familial strengths, both in terms of economic wealth and parenting style. There was no evidence that neighborhood effects on the child were mediated by the family. Rather the family modifies but does not explain the impact of neighborhoods on the child with ADHD.

We also find that improving inattention was associated with a parental history of ADHD, which runs against findings of family history as a risk factor for persistence (Franke et al., 2012). One possibility is that the association reflects contextual factors, such as a better ‘fit’ between the parent with ADHD and their affected child (Johnston et al., 2016; Psychogiou et al., 2007). Such ‘fit’ might arise through similarities in the tempo of child-parent behaviors and in greater parental, particularly maternal, understanding and acceptance of a child’s ADHD symptoms. In short, the finding highlights the need to examine families, particularly the parent- child transmission of genetic and environmental risk factors, that might drive ADHD onset and resilience factors that might promote remission.

Limitations and future directions.

Among the limitations of this study, while our assessment of neighborhood and familial contexts was broad, it omitted factors that others have tied to mental health outcomes. For example, peer groups might act as a proximal influence on a child’s mental health, including ADHD (Danzig et al., 2013; Newcomb et al., 1993; Szekely et al., 2016). Second, we conducted only a baseline assessment of neighborhoods and thus could miss how changes in the neighborhood might contribute to symptom trajectories. However, most of the neighborhood variables we considered are unlikely to change substantially during the time-frame of the study, which had a mean duration of 2.6 years. An absence of mediation of neighborhood by familial effects could also be inferred with greater confidence from longitudinal designs that included families that moved neighborhoods, or the neighborhood themselves underwent sudden changes in wealth. Additionally, the cohort had a wide age range. However, we found that the child’s age at study entry did not interact with neighborhood or family factors in the determination of the rates of symptom change. We found other child level properties including gender and symptom severity at study entry did not explain the association between neighborhood characteristics and the child’s clinical course. However, a more detailed examination of child-level factors, particularly a child’s genotype is a priority

Translational implications.

While we cannot infer mechanism from this observational study, the findings point to familial factors – particularly high degrees of familial conflict and order- as points of possible intervention. In this context, promising initial evidence is noted for interventions which target parent-child interactions to promote sensitive responsivity and decrease harsh parenting styles (Chronis-Tuscano et al., 2011). Our study suggests that such interventions may be best targeted at children living in relatively poor neighborhood that may lack the community resources to compensate for familial disadvantage.

Conclusion.

Working within a transactional model, we illustrate how the symptom trajectories of inattention are influenced by the interplay between distal, neighborhood factors and proximal, familial level processes of conflict and control. The work highlights contexts for novels interventions that might accelerate remission in this highly impairing and prevalent childhood disorder.

Supplementary Material

Supplemental

References

  1. A.P.A. (2013). Diagnostic and statistical manual of mental disorders (DSM-5®): American Psychiatric Pub. [DOI] [PubMed] [Google Scholar]
  2. Amoly E, Dadvand P, Forns J, Lopez-Vicente M, Basagana X, Julvez J, et al. (2014). Green and blue spaces and behavioral development in Barcelona schoolchildren: the BREATHE project. Environ Health Perspect, 122, 1351–1358. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bates D, Mächler M, Bolker B, & Walker S (2014). Fitting linear mixed-effects models using lme4. arXiv preprint arXiv:1406.5823. [Google Scholar]
  4. Biederman J, Milberger S, Faraone SV, Kiely K, Guite J, Mick E, et al. (1995). Impact of adversity on functioning and comorbidity in children with attention-deficit hyperactivity disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 34, 1495–1503. [DOI] [PubMed] [Google Scholar]
  5. Bronfenbrenner U (2009). The ecology of human development: Harvard university press. [Google Scholar]
  6. Burchinal M, Roberts JE, Zeisel SA, Hennon EA, & Hooper S (2006). Social Risk and Protective Child, Parenting, and Child Care Factors in Early Elementary School Years. Parenting, 6, 79–113. [Google Scholar]
  7. Butler AM, Kowalkowski M, Jones HA, & Raphael JL (2012). The relationship of reported neighborhood conditions with child mental health. Acad Pediatr, 12, 523–531. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Buu A, Dipiazza C, Wang J, Puttler LI, Fitzgerald HE, & Zucker RA (2009). Parent, family, and neighborhood effects on the development of child substance use and other psychopathology from preschool to the start of adulthood. J Stud Alcohol Drugs, 70, 489–498. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chase-Lansdale PL, & Gordon RA (1996). Economic Hardship and the Development of Five- and Six-Year-Olds: Neighborhood and Regional Perspectives. Child Development, 67, 3338–3367. [Google Scholar]
  10. Chronis-Tuscano A, O’Brien KA, Johnston C, Jones HA, Clarke TL, Raggi VL, et al. (2011). The relation between maternal ADHD symptoms & improvement in child behavior following brief behavioral parent training is mediated by change in negative parenting. Journal of Abnormal Child Psychology, 39, 1047–1057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Conway CC, Raposa EB, Hammen C, & Brennan PA (2018). Transdiagnostic pathways from early social stress to psychopathology: a 20‐year prospective study. Journal of Child Psychology and Psychiatry. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dahal S, Swahn MH, & Hayat MJ (2018). Association between Neighborhood Conditions and Mental Disorders among Children in the US: Evidence from the National Survey of Children’s Health 2011/12. Psychiatry Journal, 2018, 9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Danzig AP, Bufferd SJ, Dougherty LR, Carlson GA, Olino TM, & Klein DN (2013). Longitudinal associations between preschool psychopathology and school-age peer functioning. Child Psychiatry Hum Dev, 44, 621–632. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dearing E (2004). The developmental implications of restrictive and supportive parenting across neighborhoods and ethnicities: Exceptions are the rule. Journal of Applied Developmental Psychology, 25, 555–575. [Google Scholar]
  15. Demontis D, Walters RK, Martin J, Mattheisen M, Als TD, Agerbo E, et al. (2017). Discovery Of The First Genome-Wide Significant Risk Loci For ADHD. bioRxiv. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Derauf C, Pandey D, Liesinger JT, Ryu E, Ziegenfuss JY, & Juhn Y (2015). Cross-sectional study of perceived neighborhood collective efficacy and risk of adhd among a nationally-representative sample of children. Journal of Epidemiological Research, 2, 71. [Google Scholar]
  17. Doan SN, Fuller-Rowell TE, & Evans GW (2012). Cumulative risk and adolescent’s internalizing and externalizing problems: The mediating roles of maternal responsiveness and self-regulation. Developmental Psychology, 48, 1529. [DOI] [PubMed] [Google Scholar]
  18. Ellis B, & Nigg J (2009). Parenting practices and attention-deficit/hyperactivity disorder: new findings suggest partial specificity of effects. J Am Acad Child Adolesc Psychiatry, 48, 146–154. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Faraone SV, Biederman J, & Mick E (2006). The age-dependent decline of attention deficit hyperactivity disorder: a meta-analysis of follow-up studies. Psychological Medicine, 36, 159–165. [DOI] [PubMed] [Google Scholar]
  20. Faraone SV, & Larsson H (2018). Genetics of attention deficit hyperactivity disorder. Molecular Psychiatry, 1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Farrell AD, & Bruce SE (1997). Impact of exposure to community violence on violent behavior and emotional distress among urban adolescents. Journal of Clinical Child Psychology, 26, 2–14. [DOI] [PubMed] [Google Scholar]
  22. Flouri E, Midouhas E, Joshi H, & Tzavidis N (2015). Emotional and behavioural resilience to multiple risk exposure in early life: the role of parenting. European Child & Adolescent Psychiatry, 24, 745–755. [DOI] [PubMed] [Google Scholar]
  23. Franke B, Faraone SV, Asherson P, Buitelaar J, Bau CHD, Ramos-Quiroga JA, et al. (2012). The genetics of attention deficit/hyperactivity disorder in adults, a review. Molecular Psychiatry, 17, 960–987. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Garbarino J, & Sherman D (1980). High-risk neighborhoods and high-risk families: The human ecology of child maltreatment. Child Development, 188–198. [PubMed] [Google Scholar]
  25. Harold GT, Leve LD, Barrett D, Elam K, Neiderhiser JM, Natsuaki MN, et al. (2013). Biological and rearing mother influences on child ADHD symptoms: revisiting the developmental interface between nature and nurture. Journal of Child Psychology and Psychiatry, 54, 1038–1046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hjortebjerg D, Andersen AM, Christensen JS, Ketzel M, Raaschou-Nielsen O, Sunyer J, et al. (2016). Exposure to Road Traffic Noise and Behavioral Problems in 7-Year-Old Children: A Cohort Study. Environ Health Perspect, 124, 228–234. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Hollingshead A (1975). Four Factor Index of Social Status. New Haven: Yale University Department of Sociology. [Google Scholar]
  28. Humphrey JL, & Root ED (2017). Spatio-temporal neighborhood impacts on internalizing and externalizing behaviors in U.S. elementary school children: Effect modification by child and family socio-demographics. Social Science & Medicine, 180, 52–61. [DOI] [PubMed] [Google Scholar]
  29. Imai K, Keele L, & Yamamoto T (2010). Identification, inference and sensitivity analysis for causal mediation effects. Statistical science, 51–71. [Google Scholar]
  30. Johnston C, & Chronis-Tuscano A (2015). Families and ADHD. [Google Scholar]
  31. Johnston C, & Mash EJ (2001). Families of children with attention-deficit/hyperactivity disorder: review and recommendations for future research. Clinical child and family psychology review, 4, 183–207. [DOI] [PubMed] [Google Scholar]
  32. Johnston C, Williamson D, Noyes A, Stewart K, & Weiss MD (2016). Parent and child ADHD symptoms in relation to parental attitudes and parenting: Testing the similarity-fit hypothesis. Journal of Clinical Child & Adolescent Psychology, 1–10. [DOI] [PubMed] [Google Scholar]
  33. Kemp GN, Langer DA, & Tompson MC (2016). Childhood Mental Health: An Ecological Analysis of the Effects of Neighborhood Characteristics. J Community Psychol, 44, 962–979. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Kepley HO, & Ostrander R (2007). Family characteristics of anxious ADHD children: Preliminary results. Journal of Attention Disorders, 10, 317–323. [DOI] [PubMed] [Google Scholar]
  35. Kitzmann KM, Gaylord NK, Holt AR, & Kenny ED (2003). Child witnesses to domestic violence: a meta-analytic review. Journal of Consulting and Clinical Psychology, 71, 339. [DOI] [PubMed] [Google Scholar]
  36. Kohen DE, Leventhal T, Dahinten VS, & McIntosh CN (2008). Neighborhood disadvantage: Pathways of effects for young children. Child Development, 79, 156–169. [DOI] [PubMed] [Google Scholar]
  37. Kuntsi J, Pinto R, Price TS, van der Meere JJ, Frazier-Wood AC, & Asherson P (2014). The Separation of ADHD Inattention and Hyperactivity-Impulsivity Symptoms: Pathways from Genetic Effects to Cognitive Impairments and Symptoms. Journal of Abnormal Child Psychology, 42, 127–136. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Larsson H, Dilshad R, Lichtenstein P, & Barker ED (2011). Developmental trajectories of DSM-IV symptoms of attention-deficit/hyperactivity disorder: genetic effects, family risk and associated psychopathology. Journal of Child Psychology and Psychiatry, 52, 954–963. [DOI] [PubMed] [Google Scholar]
  39. Leventhal T, & Brooks-Gunn J (2000). The neighborhoods they live in: the effects of neighborhood residence on child and adolescent outcomes. Psychological Bulletin, 126, 309. [DOI] [PubMed] [Google Scholar]
  40. Li M, Johnson SB, Musci RJ, & Riley AW (2017). Perceived neighborhood quality, family processes, and trajectories of child and adolescent externalizing behaviors in the United States. Social Science & Medicine, 192, 152–161. [DOI] [PubMed] [Google Scholar]
  41. Lipina SJ, & Colombo JA (2009). Poverty and brain development during childhood: An approach from cognitive psychology and neuroscience: American Psychological Association. [Google Scholar]
  42. Lubke GH, Hudziak JJ, Derks EM, van Bijsterveldt TCEM, & Boomsma DI (2009). Maternal ratings of attention problems in ADHD: evidence for the existence of a continuum. Journal of the American Academy of Child & Adolescent Psychiatry, 48, 1085–1093. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Lytton H, & Romney DM (1991). Parents’ differential socialization of boys and girls: A meta-analysis. Psychological Bulletin, 109, 267–296. [Google Scholar]
  44. MacKinnon DP, Fairchild AJ, & Fritz MS (2007). Mediation analysis. Annu. Rev. Psychol, 58, 593–614. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Margolin G, & Gordis EB (2000). The effects of family and community violence on children. Annual review of psychology, 51, 445–479. [DOI] [PubMed] [Google Scholar]
  46. McKelvey LM, Whiteside-Mansell L, Bradley RH, Casey PH, Conners-Burrow NA, & Barrett KW (2011). Growing Up in Violent Communities: Do Family Conflict and Gender Moderate Impacts on Adolescents’ Psychosocial Development? Journal of Abnormal Child Psychology, 39, 95–107. [DOI] [PubMed] [Google Scholar]
  47. McLoyd VC (1998). Socioeconomic disadvantage and child development. American psychologist, 53, 185. [DOI] [PubMed] [Google Scholar]
  48. Miller LL, Gustafsson HC, Tipsord J, Song M, Nousen E, Dieckmann N, et al. (2017). Is the Association of ADHD with Socio-Economic Disadvantage Explained by Child Comorbid Externalizing Problems or Parent ADHD? Journal of Abnormal Child Psychology, 1–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Minde K, Eakin L, Hechtman L, Ochs E, Bouffard R, Greenfield B, et al. (2003). The psychosocial functioning of children and spouses of adults with ADHD. Journal of Child Psychology and Psychiatry, 44, 637–646. [DOI] [PubMed] [Google Scholar]
  50. Modesto-Lowe V, Danforth JS, & Brooks D (2008). ADHD: does parenting style matter? Clinical pediatrics, 47, 865–872. [DOI] [PubMed] [Google Scholar]
  51. Moffitt TE, Caspi A, Rutter M, & Silva PA (2001). Sex differences in antisocial behaviour: Conduct disorder, delinquency, and violence in the Dunedin Longitudinal Study: Cambridge university press. [Google Scholar]
  52. Moos RH (1990). Conceptual and empirical approaches to developing family-based assessment procedures: resolving the case of the Family Environment Scale. Fam Process, 29, 199–208; discussion 209–111. [DOI] [PubMed] [Google Scholar]
  53. Mura G, Vellante M, Nardi AE, Machado S, & Carta MG (2015). Effects of School-Based Physical Activity Interventions on Cognition and Academic Achievement: A Systematic Review. CNS Neurol Disord Drug Targets, 14, 1194–1208. [DOI] [PubMed] [Google Scholar]
  54. Murray C, & Johnston C (2006). Parenting in mothers with and without attention-deficit/hyperactivity disorder. Journal of Abnormal Psychology, 115, 52–61. [DOI] [PubMed] [Google Scholar]
  55. Newcomb AF, Bukowski WM, & Pattee L (1993). Children’s peer relations: a meta-analytic review of popular, rejected, neglected, controversial, and average sociometric status. Psychological Bulletin, 113, 99–128. [DOI] [PubMed] [Google Scholar]
  56. Nikolas M, Klump KL, & Burt SA (2012). Youth appraisals of inter-parental conflict and genetic and environmental contributions to attention-deficit hyperactivity disorder: Examination of GxE effects in a twin sample. Journal of Abnormal Child Psychology, 40, 543–554. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Odgers CL, Caspi A, Russell MA, Sampson RJ, Arseneault L, & Moffitt TE (2012). Supportive parenting mediates neighborhood socioeconomic disparities in children’s antisocial behavior from ages 5 to 12. Dev Psychopathol, 24, 705–721. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Park JL, Hudec KL, & Johnston C (2017). Parental ADHD symptoms and parenting behaviors: A meta-analytic review. Clinical Psychology Review, 56, 25–39. [DOI] [PubMed] [Google Scholar]
  59. Peris TS, & Baker BL (2000). Applications of the expressed emotion construct to young children with externalizing behavior: stability and prediction over time. J Child Psychol Psychiatry, 41, 457–462. [PubMed] [Google Scholar]
  60. Phillips DA, & Shonkoff JP (2000). From neurons to neighborhoods: The science of early childhood development: National Academies Press. [PubMed] [Google Scholar]
  61. Pingault JB, Viding E, Galera C, Greven C, Zheng Y, R., P., et al. (2015). Genetic and environmental influences on the developmental course of attention-deficit/hyperactivity disorder symptoms from childhood to adolescence. JAMA Psychiatry, doi: 10.1001/jamapsychiatry.2015.0469 T. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Plybon LE, & Kliewer W (2001). Neighborhood types and externalizing behavior in urban school-age children: Tests of direct, mediated, and moderated effects. Journal of Child and Family Studies, 10, 419–437. [Google Scholar]
  63. Polanczyk G, de Lima M, Bernardo H, Biederman J, & Rohde L (2007). The Worldwide Prevalence of ADHD: A Systematic Review and Metaregression Analysis. American Journal of Psychiatry, 164, 942–948. [DOI] [PubMed] [Google Scholar]
  64. Polderman TJC, Derks EM, Hudziak JJ, Verhulst FC, Posthuma D, & Boomsma DI (2007). Across the continuum of attention skills: a twin study of the SWAN ADHD rating scale. Journal of Child Psychology & Psychiatry & Allied Disciplines, 48, 1080–1087. [DOI] [PubMed] [Google Scholar]
  65. Pressman LJ, Loo SK, Carpenter EM, Asarnow JR, Lynn D, McCracken JT, et al. (2006). Relationship of Family Environment and Parental Psychiatric Diagnosis to Impairment in ADHD. Journal of the American Academy of Child & Adolescent Psychiatry, 45, 346–354. [DOI] [PubMed] [Google Scholar]
  66. Psychogiou L, Daley D, Thompson M, & Sonuga‐Barke E (2007). Testing the interactive effect of parent and child ADHD on parenting in mothers and fathers: A further test of the similarity‐fit hypothesis. British Journal of Developmental Psychology, 25, 419–433. [Google Scholar]
  67. Razani N, Hilton JF, Halpern-Felsher BL, Okumura MJ, Morrell HE, & Yen IH (2015). Neighborhood characteristics and ADHD: Results of a National Study. Journal of Attention Disorders, 19, 731–740. [DOI] [PubMed] [Google Scholar]
  68. Reich W (2000). Diagnostic interview for children and adolescents (DICA). Journal of the American Academy of Child & Adolescent Psychiatry, 39, 59–66. [DOI] [PubMed] [Google Scholar]
  69. Rocha V, Ribeiro AI, Severo M, Barros H, & Fraga S (2017). Neighbourhood socioeconomic deprivation and health-related quality of life: A multilevel analysis. PloS one, 12, e0188736. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Russell AE, Ford T, Williams R, & Russell G (2016). The association between socioeconomic disadvantage and attention deficit/hyperactivity disorder (ADHD): a systematic review. Child Psychiatry & Human Development, 47, 440–458. [DOI] [PubMed] [Google Scholar]
  71. Selner-O’Hagan MB, Kindlon DJ, Buka SL, Raudenbush SW, & Earls FJ (1998). Assessing Exposure to Violence in Urban Youth. The Journal of Child Psychology and Psychiatry and Allied Disciplines, 39, 215–224. [PubMed] [Google Scholar]
  72. Silk JS, Sessa FM, Sheffield Morris A, Steinberg L, & Avenevoli S (2004). Neighborhood Cohesion as a Buffer Against Hostile Maternal Parenting. Journal of Family Psychology, 18, 135–146. [DOI] [PubMed] [Google Scholar]
  73. Sundquist J, Li X, Ohlsson H, Råstam M, Winkleby M, Sundquist K, et al. (2015). Familial and neighborhood effects on psychiatric disorders in childhood and adolescence. Journal of Psychiatric Research, 66–67, 7–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  74. Szekely E, Pappa I, Wilson JD, Bhamidi S, Jaddoe VW, Verhulst FC, et al. (2016). Childhood peer network characteristics: genetic influences and links with early mental health trajectories. J Child Psychol Psychiatry, 57, 687–694. [DOI] [PubMed] [Google Scholar]
  75. Taylor AF, & Kuo FE (2009). Children with attention deficits concentrate better after walk in the park. J Atten Disord, 12, 402–409. [DOI] [PubMed] [Google Scholar]
  76. Vilhjalmsdottir A, Gardarsdottir RB, Bernburg JG, & Sigfusdottir ID (2016). Neighborhood income inequality, social capital and emotional distress among adolescents: A population-based study. J Adolesc, 51, 92–102. [DOI] [PubMed] [Google Scholar]
  77. Walton E, Pingault J-B, Cecil CA, Gaunt TR, Relton C, Mill J, et al. (2017). Epigenetic profiling of ADHD symptoms trajectories: a prospective, methylome-wide study. Molecular Psychiatry, 22, 250. [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Ward MF, Wender PH, & Reimherr FW (1993). The Wender Utah Rating Scale: an aid in the retrospective diagnosis of childhood attention deficit hyperactivity disorder.[erratum appears in Am J Psychiatry 1993 Aug;150(8):1280]. American Journal of Psychiatry, 150, 885–890. [DOI] [PubMed] [Google Scholar]
  79. Weiler MD, Bernstein JH, Bellinger DC, & Waber DP (2000). Processing Speed in Children With Attention Deficit/Hyperactivity Disorder, Inattentive Type. Child Neuropsychology, 6, 218–234. [DOI] [PubMed] [Google Scholar]
  80. Willcutt EG, Pennington BF, Olson RK, & DeFries JC (2007). Understanding comorbidity: A twin study of reading disability and attention-deficit/hyperactivity disorder. American Journal of Medical Genetics Part B: Neuropsychiatric Genetics, 144B, 709–714. [DOI] [PubMed] [Google Scholar]
  81. Wood AC, Rijsdijk F, Asherson P, & Kuntsi J (2009). Hyperactive-Impulsive Symptom Scores and Oppositional Behaviours Reflect Alternate Manifestations of a Single Liability. Behavior Genetics, 39, 447–460. [DOI] [PMC free article] [PubMed] [Google Scholar]
  82. Xue Y, Leventhal T, Brooks-Gunn J, & Earls FJ (2005). Neighborhood residence and mental health problems of 5-to 11-year-olds. Archives of General Psychiatry, 62, 554–563. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental
NIHMS1042312-supplement-Supplemental.docx (34.9KB, docx)

RESOURCES