Social Inequality and Cultural Diversity

Article

Journal of Family Issues 2023, Vol. 44(2) 386–408 © The Author(s) 2021 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/0192513X211048477 journals.sagepub.com/home/jfi

Internalizing Symptoms in Children Being Reared by Grandparents in Rural Appalachia: Risk and Protective Factors

Peggy S. Keller1, Shuang Bi1, and Nancy Schoenberg1

Abstract The present study investigates factors associated with anxiety, depression, and stress in children being reared by their grandparents in rural Appalachia. Grandparent-headed households, in which grandparents have primary respon- sibility for care of children, are increasing in number. However, research is needed on child risk for internalizing symptoms in this context. Participants included 35 children aged 5–18 years and one of their custodial grandparents. Interviewers read questionnaires to grandparents and children assessing family functioning and child mental health and participants indicated their answers. Mean scores for child internalizing symptoms and stress were on the higher end of the scale of measurement. Lower grandparent positive parenting, grandparent not having formal custody of children, greater grandparent mental health issues, lower grandparent education, and lower financial status were correlated with greater child anxiety, depression, and stress. Results also indicated that correlates of children’s internalizing symptoms differed based on child sex and age.

Keywords grandparents, children, Appalachia, internalizing, anxiety, depression

1University of Kentucky, Lexington, KY, USA

Corresponding Author: Peggy S. Keller, Department of Psychology, University of Kentucky, 012i Kastle Hall, Lexington, KY 40506, USA. Email: [email protected]

Grandparent-headed households, in which grandparents have the primary responsibility for the care of children (“grandfamilies”), are one of the fastest growing family types in the United States and around the world (Glaser et al., 2018). Grandfamilies have become more common for a variety of reasons, including parental death, parental incarceration, parental addiction, and pa- rental abandonment of children (Downie, Hay, Horner, Wichmann, & Hislop, 2010). Research on this population has focused on the physical and mental health of grandparents. This research emphasizes the high levels of stress, poor economic conditions, and other factors faced by grandfamilies that place grandparent well-being at risk (e.g., Yoo & Russell, 2020). The purpose of the present study is to address the gap in research on child well-being in grandfamilies. Specifically, this study explores factors that are associated with anxiety, depression, and stress in children being reared by their grandparents in rural Appalachia. Factors selected for the study are those that have been identified as characteristic of the region and are linked to internalizing symptoms generally.

There has been no research on the psychological health of children being reared by grandparents in Appalachia, but there are a number of studies from other geographical areas. Most of this research focuses on externalizing symptoms (e.g., Goulette, Evans, & King, 2016; Pilkauskas & Dunifon, 2016). Research indicates that children reared by grandparents have better psychological outcomes than children placed in foster care (Rubin et al., 2008), but that children may fare worse in grandfamilies than in traditional two-parent biological families who are also economically disadvantaged (Pittman&Boswell, 2007). There have been a few studies that examined internalizing difficulties in children reared by grandparents. One study of 20 Australian children indicated that although many children in grandfamilies report being well-adjusted, some children express problems with anxiety (Downie et al., 2010). Among a sample of Chinese adolescents being reared by their grandparents when their parents migrated out of the region for work, rates of anxiety and depression were significantly higher compared to their parent-reared peers (Tang et al., 2018). These findings suggest that children being raised by grandparents may be at risk for stress, anxiety and depression, but additional research is needed.

The present study focuses on two rural central Appalachian counties in Kentucky. This part of Appalachia is economically depressed, with high concentrations of poverty and low levels of education (Jenkins-Howard, Stephenson, & Mains, 2013). Rates of drug addiction are very high; pre- scription opioid addiction is a particularly serious problem (Schalkoff et al., 2020). Drug involvement is associated with criminal activity and incarceration (Roberson, Cortez, Trull, & Lenger, 2020). Further, Kentucky has the second highest rate of child abuse in the United States (Administration for Children and Families, 2014). Thus, many of the reasons for parental inability to care for children are prevalent in the rural Appalachian region of Kentucky. In fact,

Keller et al. 387

nearly one in five households is led by grandparents in the study counties, a number that is twice the national average (Pollard & Jacobsen, 2014). Parental financial concerns and low education levels are both risk factors for child internalizing symptoms (Masarik & Conger, 2017) and are therefore included as risk factors in the current study. Indeed, during the recent COVID-19 pandemic, financial hardship was associated with greater grandparent par- enting stress indirectly through greater grandparent mental health struggles (Xu, Wu, Levkoff, & Jedwab, 2020).

Although Appalachian grandfamilies may experience several disadvan- tages, they also have numerous assets. These assets include strong family ties, commitment to providing social support, and long standing, well-trusted organizations such as faith groups (Schoenberg, Hatcher, & Dignan, 2008). Family cohesion and support among family members are high priorities, and families provide a sense of identity and belonging (Lancki et al., 2018). Family cohesion is associated with lower anxiety and depression, less suicide ideation, and fewer suicide attempts among various cultural groups (Park, Unutzer, & Grembowski, 2014; Rapp, Lau, & Chavira, 2016; White, Shelton, & Elgar, 2014). Spirituality is also a major part of life in Appalachian regions, and may encompass living in harmony with others and having a strong re- lationship with God or other divine presence (Dunfee, Brown, & Schoenberg, 2021). Spirituality and the religious coping are consistently related to better physical and mental health (e.g., Whitehead & Bergeman, 2019). Thus, understanding Appalachian grandfamilies requires consideration of their strengths as well as their challenges (DeFrain, 2014), and grandparent social support and religiosity are included as potential protective factors.

Another key factor for child internalizing symptoms in grandfamilies is likely to be grandparent well-being, which in turn may contribute to family conflict and poor grandparent parenting (“grandparenting”). Grandparents may find rearing their grandchildren to be highly stressful, as it is an un- expected role at their stage of life, their grandchildren may have needs and difficulties that are difficult for them to address, and taking over care of grandchildren can result in social isolation (Hayslip, Fruhauf, & Fish, 2021; Waldrop & Weber, 2001). More time caring for grandchildren is associated with greater risk for depression in grandparents (Brunello & Rocco, 2018). US grandparent psychological distress (a combination of anxiety and depression symptoms) is related to greater marital stress and dysfunctional parenting (a combination of parenting stress and poor parenting practices), which in turn are related to children’s greater internalizing symptoms; more social support, better health, and higher income were associated with lower grandparent distress (Smith & Hancock, 2010; Smith, Palmieri, Hancock, & Richardson, 2008). In a study of predominantly African American grandfamilies, grandparent psychological distress (based on a global measure of mental health–related stress), a home environment providing less support to children,

388 Journal of Family Issues 44(2)

and low family resources were related to greater behavioral problems in children (Kelley, Whitley, & Campos, 2011). Grandparent care and over- protection was associated with greater child internalizing and externalizing symptoms in a sample of intact Chinese families, controlling for mother and father care and overprotection (Li, Cui, Kok, Deatrick, & Liu, 2019).

Given the lack of research on grandfamilies in the Appalachian context and the greater constellation of needs in this environment, we sought to understand associations between grandfamily characteristics and child internalizing symptoms in rural Appalachia. Specifically, we examine strengths and lim- itations of Appalachian grandfamilies. We hypothesize that possible weak- nesses such as less education, lower financial status, greater grandparent depression and stress, will be associated with children’s greater stress, anxiety and depression. We also hypothesize that possible strengths, such as grandparent social support, grandparent positive parenting, and grandparent religiosity, will be associated with children’s lower stress, anxiety and de- pression. Finally, we examine differential associations based on child sex and child age. There has been little to no research on differential risk for ad- justment problems in grandfamilies, and thus these questions are exploratory.

Method

Participants

Participants were 35 children (16 girls) between the ages of 5 and 18 years (M = 12.56 years, SD = 3.68) and one of their grandparents (25 women, 19 married or living with a partner). Families were recruited via convenience sampling and snowball sampling. Recruitment was conducted by the local project manager, who attended grandparent coalition meetings or other community gatherings to recruit participants. Eligibility criteria were: (1) grandparent was at least 55 years of age; (2) grandparent served as a primary caregiver to at least one grandchild; (3) grandchild is between the ages of six and 18 years. Only one child (chosen by grandparent if multiple were eligible) and one grandparent per grandfamily participated. Twenty-three grandparents had formal legal custody of the grandchild, 11 grandparents did not have formal legal custody but took care of the grandchild most of the time, and one grandparent took care of the grandchild about half of the time. In the majority of cases, grandparents took over care of the children due to parental desertion or neglect (61% of biological mothers and 65% of biological fathers) or substance use problems (26% of mothers and 17% of fathers). In the re- maining cases, biological parents had died (9% of mothers and 13% of fathers) or were incarcerated (1 father). Grandparent age ranged from 57 to 85 years,M = 67.65, SD = 5.19. Consistent with the demographics of the region, 100% of the participants were white (96.5% of residents in this area are white based on

Keller et al. 389

census data; https://datausa.io/profile/geo/lee-county-ky/#intro), only 13 grandparents had a high school diploma or more education.

Procedure

This study was conducted with the approval of the university Institutional Review Board; informed consent was obtained from grandparents and in- formed assent was obtained from children. Research staff visited the homes of the grandfamilies and separated the grandparent and grandchild into different rooms for interviews. All questions were asked in interview format to avoid issues with literacy. Participants were provided with cards listing the response options on the Likert scales to improve the ease of answering the questions. Interviews lasted approximately 1 hour, and experimenters were instructed to inquire whether participants would like a break 20 minutes into the interview (no participants requested one) and to monitor for participant fatigue and suggest a break if fatigue became apparent (no participants demonstrated fatigue and instead most appeared to enjoy the interview).

Measures

Child anxiety. Children completed the Penn State Worry Questionnaire adapted for use with children (Chorpita, Tracey, Brown, Collica, & Barlow, 1997). The 14 items are rated on a scale from 1 to 4, with higher scores indicating greater worry; Cronbach’s α in the current sample was excellent, .92.

Child depression. Children completed the Center for Epidemiologic Studies Depression scale for Children (Fendrich, Weissman, &Warner, 1990). The 20 items are rated on a four-point scale, with higher scores indicating greater frequency of the symptom; Cronbach’s α in the current sample was very good, .85. Item scores are summed to provide a total score.

Child stress. Children completed the first portion of the Responses to Stress Scale (Family Stress Version) (Connor-Smith, Compas, Wadsworth, Thom- sen, & Saltzman, 2000). This portion includes 11 items representing different family stress experiences (e.g., arguing with grandparents and arguing with siblings) rated on a scale from one to four in terms of how much each of the experiences stresses them. Scores were computed by averaging responses across the 11 items. Cronbach’s α in the current sample was acceptable, .77.

Grandparent health. Grandparents completed the Short Form-36 (SF-36) Health Survey. Items pertaining to ability to walk one or multiple blocks were modified to “one hundred” or “several hundred yards” given the rural setting. The survey was scored using the SF-36 scoring manual (Ware,

390 Journal of Family Issues 44(2)

Kosinski, & Keller, 1994). The Physical Health and Mental Health Scores were computed. Higher scores indicate better health. In the general US population, each component has a norm ofM = 50, SD = 10. In the age group of 65–74 years, norms areM = 43.65 SD = 11.02 andM = 52, SD = 9.53 for the physical and mental health components, respectively.

Grandparent depression. Grandparents completed the Patient Health Questionnaire-9 (Martin, Rief, Klaiberg, & Braehler, 2006). The nine items are rated on a four-point scale, with higher scores indicating greater de- pression. Reliability was good in the current sample, Cronbach’s α = .80. For older adults, a score of six or higher is indicative of any depressive disorder (Lamers et al., 2008). Based on these criteria, 11 grandparents (31.4% of the sample) may have a depressive disorder.

Grandparent parenting stress. Grandparent parenting stress was assessed with a version of the Parenting Stress Scale (Berry & Jones, 1995) modified to use “grandparent” instead of “parent”. Items are rated on a scale from 1 to 5, with higher scores indicating greater agreement with each statement. Of the 18 items, eight reflect positive themes of parenthood and are reverse scored and added to the scores for the other 10 items. Cronbach’s α was good in the current sample, .79.

Grandparent positive parenting. Children completed the Acceptance subscale of the Child Report of Parental Behavior Inventory (Schaefer, 1965). Children rated how much their grandparents make them feel accepted in 10 different ways (e.g., “smile at me very often”) on a scale from one to 3. Higher scores indicate greater acceptance of the child. Cronbach’s α was excellent in the current sample, .91.

Grandparent religiosity. Grandparents answered two questions: (1) How often do you attend church or other religious meetings? And (2) How often do you spend time in private religious activities, such as prayer, meditation, or Bible study? Both questions were rated on a six-point scale, which higher scores indicating greater frequency of religious activity. Given that public and private religious behaviors are conceptually distinct (e.g., Nonnemaker, McNeely, & Blum, 2003), the two questions were treated separately for analysis.

Grandparent social support. Grandparents completed the Medical Outcomes Study Social Support Survey (Sherbourne & Stewart, 1991). This survey includes four subscales for the different types of social support grandparents may experience: emotional/informational support, tangible support, affec- tionate support, and positive social interaction. There are 18 items, rated on a five-point scale in terms of how frequently a person is available to provide

Keller et al. 391

support. Scores were computed by averaging item responses for each scale; Cronbach’s α was excellent in the current sample, ranging from .93 to .98.

Grandfamily demographics. We considered the following demographic variables: Grandparent age, child age, child sex, marital status (partnered or single grand- parent), number of children in the household, highest level of grandparent edu- cation, and perceived financial status. Perceived financial status was rated on a scale from 1 to 3, with 1 reflecting “I have more than I need to live well” (n = 3), 2 indicating “I have just about enough to get by” (n = 15), and 3 indicating “I sometimes struggle to make ends meet” (n = 17). To evaluate differences in correlations due to child age, the sample was divided into children 12 and younger, and children older than 12 years, resulting in 17 children in each age group. Child age was otherwise treated as a continuous variable (measured in years).

Results

Child Internalizing Problems

Means and SDs for variables are provided in Table 1. Large SDs indicate great variability in child depression and anxiety symptoms. For the measure of child depression, scores of 15 or higher indicate potentially clinical levels of depression and 19 children (57.6%) met this criterion. This value is substantially higher than the potential clinical depression prevalence of 24.7% of children and adolescents who have been exposed to trauma (e.g., natural disaster and maltreatment; (Vibhakar, et al., 2019) and the 17% of girls and 7% of boys who are involved in the US child welfare system (Orton et al., 2009). Mean levels of anxiety were also fairly high. On average, children reported relatively low stress. The mean score, 1.79 is just below the rating of “a little” stressed. As shown in Table 2, mean ratings on stress items differed little from “not at all” stressed to “a little” stressed. However, over 21% of children indicated that they were “very” stressed by competing with siblings, not being as close to siblings as they would like, and siblings breaking or taking belongings. Given this pattern, a mean score for sibling-related items (M = 2.06, SD = 0.92) and a separate mean score for grandparent-related items (M = 1.67, SD = 0.56) was computed. These means were significantly different, t (31) = 2.47, p < .05. Thus, children appear to be more stressed by sibling relationships than relationships with their grandparents.

Grandfamily Strengths and Weaknesses

Means (Table 1) show that, in some ways, grandparents were doing well. The mean score for grandparent depression on the Patient Health Questionnaire was below the clinical cut-off, positive parenting had a mean score close to the top of the possible range, and the means of the social support measures were close to 4,

392 Journal of Family Issues 44(2)

indicating that social support is available “most of the time.” However, the mean perceived financial status was halfway between the option “I have just enough to get by” and the worst option, “I sometimes struggle to make ends meet.” Ed- ucation levels were fairly low, being only a 10th grade education on average. Mean scores for grandparenting stress were near the middle of the scales. The mean score for grandparent physical health on the Short Form-36 is 1.35 SDs below the mean of a general national population and 0.77 SDs below the mean of the corresponding age (65–74 years) population. The mental health mean score on the Short Form-36 is 0.42 SDs below the national mean and 0.63 SD below the corresponding age group mean.

Correlations Among the Whole Sample

See Table 3. Greater grandparent parenting stress was associated with greater child depression symptoms, r = .39, p < .05. Greater grandparent positive parenting was related to lower child depression symptoms, r = �.48, p < .05.

Table 1. Means and Standard Deviations for Study Variables.

Variable M SD

Child depression (scale = 0–60) 15.00 11.04 Child anxiety (scale = 14–56) 28.55 10.31 Child stress (scale = 1–4) 1.79 0.57 Grandparent age (years) 67.65 5.19 Child age (years) 12.56 3.68 Child female 45.7% Grandparent single 45.7% Number of children <18 years in home 1.86 1.12 Grandparent has formal custody 66% Grandparent education (grade level attained) 10.29 2.93 Poor perceived financial status (scale = 1–3) 2.40 0.65 Religious service attendance (scale = 1–6) 3.74 1.88 Private religious activity (scale = 1–6) 4.00 1.78 Grandparent physical health 35.14 11.19 Grandparent mental health 46.04 8.91 Grandparent depression (scale = 0–27) 4.20 4.26 Grandparent parenting stress (scale = 18–90) 30.74 8.08 Grandparent social support (scale = 1–5) Emotional/informational support 3.74 1.31 Tangible support 4.05 1.14 Affectionate support 4.23 1.14 Positive social interaction 3.81 1.45 Grandparent positive parenting (scale = 1–3) 2.67 0.46

Keller et al. 393

A greater number of children under the age of 18, greater grandparent de- pression symptoms, and being a single grandparent were all associated with marginally or significantly greater child anxiety, rs = .33 to .37. A follow up t test indicated that anxiety symptoms were marginally greater for children when grandparents were single (M = 32.00, SD = 11.79) rather than partnered (M = 25.29; SD = 7.70), t (31) = 1.95, p = .061. Grandparent custody of children and positive parenting were associated with children’s lower stress levels, r =�.43 and�.42, ps < .05, respectively. Child stress was significantly greater for children when grandparents did not have custody (M = 2.11, SD = 0.43) compared to when they did have custody (M = 1.61; SD = 0.65), t (31) = 2.68, p < .05.

Given the potential differences in sibling-related stress and grandparent- related stress, additional correlations were examined with these separate scores. Some differences were noted between the two types of stress. Positive parenting was associated with lower grandparent-related stress, r = �.61, p < .001, but not sibling-related stress, r = �.05, p = .77. Grandparents having custody of children was related to sibling-related stress, r =�.53, p < .01, but not to grandparent-related stress, r = �.25, p = .17. A follow up analysis

Table 2. Sources of Child Stress.

How stressful have each of the following things been for you in the last 6 months? M SD Range

% Somewhat stressed

% Very stressed

Arguing with your grandparents 1.76 0.83 1–4 6.1 6.1 Your grandparents arguing with each other

1.52 0.78 1–4 6.9 3.4

Competing with your brother or sister 2.17 1.27 1–4 21.9 21.9 Your grandparents not understanding you 1.94 1.06 1–4 15.2 12.1 Having a hard time talking with your grandparents

1.64 1.03 1–4 3.0 12.1

Your grandparents hassling or nagging you 1.76 1.06 1–4 9.1 12.1 Arguing or fighting with your brother/ sister(s)

2.17 1.17 1–4 18.8 18.8

Not being as close to your brother/ sister(s) as you’d like

1.97 1.27 1–4 9.4 21.9

Your brother/sister(s) messing up, breaking, or taking your belongings

1.97 1.27 1–4 9.4 21.9

Not spending as much time as you would like to with your grandparents

1.70 0.92 1–4 21.3 3.0

Having other kinds of problems with your grandparents

1.21 0.60 1–3 9.1 0

Note. Scale is (1) “not at all”; (2) “a little”; (3) “somewhat”; and (4) “very” stressed.

394 Journal of Family Issues 44(2)

indicated that sibling-related stress was significantly greater when grand- parents did not have custody (M = 2.73, SD = 0.93) compared to when they did (M = 1.71, SD = 0.70), t (30) = 3.46, p < .01. Grandparent education was

Table 3. Correlations between Grandfamily Characteristics and Child Internalizing Symptoms.

Variable Child

depression Child anxiety

Child stress

Child grandparent-

related stress

Child sibling- related stress

Grandparent age �.20 .08 .13 .10 .10 Child age �.09 .19 �.01 .13 �.20 Child female .08 .18 .02 �.02 .10 Grandparent single �.02 .33+ .11 .02 .20 Number of children

<18 years in home �.13 .35* �.03 �.11 .05

Grandparent has formal custody

.01 .00 �.43* .25 �.53**

Grandparent education �.12 �.20 �.27 .02 �.42* Poor perceived financial

status .01 .35* .10 .03 .17

Religious service attendance

.00 �.09 .01 .07 �.05

Private religious activity �.06 �.14 .10 .00 �.20 Grandparent physical

health �.07 �.18 �.10 �.08 �.17

Grandparent mental health

�.29 �.19 �.20 �.18 �.09

Grandparent depression .24 .37* .23 .17 .25 Grandparent parenting

stress .39* .19 .10 .20 �.06

Grandparent social support Emotional/ informational support

�.23 .05 .01 �.16 .19

Tangible support .05 �.01 .27 .12 .34 Affectionate support �.22 �.20 �.04 �.23 .14 Positive social interaction

�.21 �.14 �.05 �.27 .20

Grandparent positive parenting

�.48* �.43* �.42* �.06 .07

Note. +p < .10, *p < .05.

Keller et al. 395

related to child sibling-related stress, r = �.42, p < .05, but not to child grandparent-related stress, r = �.02, p = .90.

Differences Based on Child Sex

No child sex differences on child depression, anxiety, overall stress, grandparent-related stress, or sibling-related stress were observed. Similarly, no gender differences in grandfamily strengths or weaknesses were observed. Finally, correlations were estimated separately for boys and girls (Table 4). Several associations were observed, 14 for boys and 12 for girls. Of these, only two associations for boys were for the same variables as associations for girls (these were associations between positive parenting and child anxiety and child stress). This suggests an overall different pattern of relations for boys and girls. For example, grandparent custody of children was related to lower stress for boys but not for girls. Similarly, grandparent education was related to lower stress for boys but not for girls. In contrast, grandparent mental health, lower depression, and lower parenting stress were each related to lower in- ternalizing symptoms in girls but not in boys. Although the number of children less than 18 years of age in the home, poor financial status, and lower grandparent religiosity were all associated with lower anxiety symptoms for boys, it was grandparent younger age that was associated with lower anxiety for girls. Overall, it appeared that many associations between grandfamily characteristics and internalizing symptoms for boys were for demographic characteristics. In contrast, many of these associations for girls included grandparent mental health.

Differences Based on Child Age

There were no associations between child age and child internalizing symptoms or grandfamily characteristics. To determine whether correlations between study variables were different based on child age, the sample was divided into children aged 12 years and younger (“younger children”) and children older than 12 years (“older children”). Each group included 17 children. Correlations then were estimated separately for each group (Table 5). Several correlations were observed, 13 for younger children and 17 for older children. Among these, only one association for younger children involved the same variables as an association for older children (the association between positive parenting and lower child depression symptoms). This indicates that associations are generally different for younger versus older children. For example, grandparent social support was consistently related to lower anxiety in younger children, but this was less the case for older children. Similarly, religiosity was only associated with younger children’s internalizing symp- toms. Curiously, greater religiosity was associated with more child stress. In

396 Journal of Family Issues 44(2)

T ab

le 4.

C or re la tio

ns be tw

ee n G ra nd

fa m ily

C ha ra ct er is tic s an d Bo

ys an d G ir ls In te rn al iz in g Sy m pt om

s.

Bo ys

(N = 19

) G ir ls (N

= 16

)

V ar ia bl e

Bo ys

de pr es si on

Bo ys

an xi et y

Bo ys

st re ss

G ra nd

pa re nt -

re la te d st re ss

Si bl in g- re la te d

st re ss

G ir ls

de pr es si on

G ir ls

an xi et y

G ir ls

st re ss

G ra nd

pa re nt -

re la te d st re ss

Si bl in g- re la te d

st re ss

G ra nd

pa re nt

ag e

�. 19

�. 23

.4 4+

.4 0

.2 6

�. 18

.4 6+

�. 07

�. 17

.0 5

C hi ld

ag e

�. 27

.2 5

�. 28

.0 2

�. 51

* .0 8

.1 2

.2 3

.2 5

.0 8

G ra nd

pa re nt

si ng le

�. 16

.2 1

�. 04

�. 10

.0 5

.1 2

.4 6

.2 4

.1 7

.3 4

N um

be r of

ch ild re n

�. 02

.5 3*

.1 5

.0 4

.2 0

�. 26

.0 4

�. 24

�. 33

�. 14

G ra nd

pa re nt

ha s

cu st od

y �. 05

.0 3

�. 76

* �. 47

* �. 74

* .0 9

.0 7

�. 16

�. 03

�. 33

G ra nd

pa re nt

ed uc at io n

�. 20

�. 20

�. 43

+ �. 13

�. 58

* �. 01

�. 10

�. 15

.0 6

�. 28

Po or

fi na nc ia ls ta tu s

.3 3

.4 4+

.3 7

.3 7

.1 9

�. 32

.1 3

�. 16

�. 35

.1 5

R el ig io us

se rv ic e

at te nd

�. 12

�. 41

+ �. 14

�. 18

�. 05

.1 6

.4 3

.1 3

.0 4

.1 9

Pr iv at e re lig io us

ac tiv ity

�. 24

�. 53

* �. 03

�. 07

.0 2

.1 2

.3 3

.2 2

.0 6

.3 8

G ra nd

pa re nt

ph ys

he al th

�. 08

�. 24

�. 28

�. 23

�. 21

�. 02

�. 02

.0 8

.0 9

�. 10

G ra nd

pa re nt

m en t

he al th

�. 18

�. 17

.2 9

.2 0

.2 6

�. 40

�. 28

�. 61

* �. 57

* �. 42

G ra nd

pa re nt

de pr es si on

�. 08

.3 4

�. 08

�. 09

�. 03

.5 6*

.3 4

.5 5*

.5 1+

.5 2+

(c on tin ue d)

Keller et al. 397

T ab

le 4.

(c on

tin ue d)

Bo ys

(N = 19

) G ir ls (N

= 16

)

V ar ia bl e

Bo ys

de pr es si on

Bo ys

an xi et y

Bo ys

st re ss

G ra nd

pa re nt -

re la te d st re ss

Si bl in g- re la te d

st re ss

G ir ls

de pr es si on

G ir ls

an xi et y

G ir ls

st re ss

G ra nd

pa re nt -

re la te d st re ss

Si bl in g- re la te d

st re ss

G ra nd

pa re nt

pa re nt

st re ss

.2 9

.2 5

�. 13

.1 2

�. 37

.5 7*

.1 2

.4 7+

.3 6

.4 2

G ra nd

pa re nt

su pp

or t

Em ot io na l/

in fo rm

at io na l

.0 1

.1 3

.1 6

.0 2

.2 4

�. 34

.0 0

�. 04

�. 28

.1 9

T an gi bl e su pp

or t

.1 9

�. 09

.3 9

.3 1

.2 1

�. 08

.0 8

.2 0

�. 05

.4 4

A ffe ct io na te

su pp

or t

�. 18

.0 0

�. 02

�. 14

.1 8

�. 24

�. 35

�. 05

�. 29

.1 2

Po si tiv e

in te ra ct io n

�. 19

.0 4

�. 03

�. 24

.3 2

�. 25

�. 38

�. 08

�. 30

.1 1

Po si tiv e pa re nt in g

�. 60

* �. 42

+ �. 30

�. 63

* .2 2

�. 38

�. 48

+ �. 61

* �. 60

* �. 46

N ot e. +p

< .1 0,

*p < .0 5.

398 Journal of Family Issues 44(2)

T ab

le 5.

C or re la tio

ns be tw

ee n G ra nd

fa m ily

C ha ra ct er is tic s an d O ld er

an d Y ou

ng er

C hi ld re n’ s In te rn al iz in g Sy m pt om

s.

Y ou

ng er

ch ild re n (N

= 17

) O ld er

ch ild re n (N

= 17

)

V ar ia bl e

Y ou

ng er

ch ild

de pr es si on

Y ou

ng er

ch ild

an xi et y

Y ou

ng er

ch ild

st re ss

G ra nd

pa re nt -

re la te d

st re ss

Si bl in g-

re la te d

st re ss

O ld er

ch ild

de pr es si on

O ld er

ch ild

an xi et y

O ld er

ch ild

st re ss

G ra nd

pa re nt -

re la te d

st re ss

Si bl in g-

re la te d

st re ss

G ra nd

pa re nt

ag e

�. 30

.3 7

.1 9

.1 1

.1 7

�. 16

�. 13

.0 2

.1 3

�. 11

C hi ld

ag e

�. 63

* .3 5

�. 42

�. 39

�. 28

�. 12

�. 34

�. 18

�. 01

�. 31

G ra nd

pa re nt

si ng le

�. 13

.6 1*

�. 24

�. 11

�. 20

.0 7

.1 7

.3 1

.1 3

.4 1+

Fe m al e ch ild

�. 13

.3 2

�. 37

�. 19

�. 33

.2 6

.1 1

.3 4

.1 5

.4 5+

N um

be r of

ch ild re n

�. 37

.2 0

�. 16

�. 18

�. 17

.0 4

.5 8*

.0 4

�. 02

.1 0

G ra nd

pa re nt

cu st od

y .1 6

.2 7

�. 17

�. 17

�. 22

�. 09

�. 23

�. 58

* �. 29

�. 71

*

G ra nd

pa re nt

ed uc at io n

�. 31

�. 62

* �. 47

+ �. 30

�. 41

.0 3

.2 3

�. 13

.1 7

�. 42

+

Po or

fi na nc ia ls ta tu s

�. 01

.6 1*

�. 11

�. 28

.1 9

.0 1

.0 8

.1 9

.2 1

.1 1

R el .S er vi ce

at te nd

.1 0

�. 10

.4 7+

.3 1

.4 2

�. 07

�. 04

�. 35

�. 32

�. 27

Pr iv at e re l. A ct iv ity

.1 0

.0 9

.4 0

.1 5

.5 5*

�. 20

�. 41

�. 15

�. 16

�. 08

G ra nd

pa re nt

ph ys

he al th

�. 17

�. 48

+ �. 11

�. 01

�. 35

�. 03

.0 7

�. 17

�. 22

�. 05

G ra nd

pa re nt

m en t

he al th

�. 03

.2 3

.1 3

.2 6

.1 0

�. 42

+ �.4

3+ �. 34

�. 33

�. 24

G ra nd

pa re nt

de pr es si on

�. 07

.0 8

�. 14

�. 20

.0 9

.4 3+

.5 6*

.4 7+

.3 6

.4 3+

G ra nd

pa re nt

st re ss

.2 1

.1 0

�. 26

�. 06

�. 44

.5 2*

.2 6

.3 2

.3 6

.1 6

(c on tin ue d)

Keller et al. 399

T ab

le 5.

(c on

tin ue d)

Y ou

ng er

ch ild re n (N

= 17

) O ld er

ch ild re n (N

= 17

)

V ar ia bl e

Y ou

ng er

ch ild

de pr es si on

Y ou

ng er

ch ild

an xi et y

Y ou

ng er

ch ild

st re ss

G ra nd

pa re nt -

re la te d

st re ss

Si bl in g-

re la te d

st re ss

O ld er

ch ild

de pr es si on

O ld er

ch ild

an xi et y

O ld er

ch ild

st re ss

G ra nd

pa re nt -

re la te d

st re ss

Si bl in g-

re la te d

st re ss

G ra nd

pa re nt

su pp

or t

Em ot io na l/

in fo rm

at io na l

�. 32

�. 57

�. 08

�. 30

.1 0

�. 17

.8 0*

.1 3

�. 03

.2 9

T an gi bl e su pp

or t

�. 34

�. 62

+ �. 05

�. 25

.1 8

.3 1

.5 0

.5 1

.4 1

.4 5

A ffe ct io na te

su pp

or t

�. 39

�. 73

* �. 11

�. 38

.1 3

�. 16

.2 8

�. 05

�. 24

.2 1

Po si tiv e

in te ra ct io n

�. 30

�. 65

+ �. 05

�. 25

.1 3

�. 19

.3 0

�. 08

�. 36

.2 9

Po si tiv e pa re nt in g

�. 46

+ �. 56

* �. 34

�. 28

�. 28

�. 53

* �. 37

�. 47

+ �. 69

* �. 07

N ot e. Y ou

ng er

ch ild re n ar e 12

ye ar s of

ag e or

yo un ge r, an d ol de r ch ild re n ar e ol de r th an

12 ye ar s of

ag e;

+ p < .1 0,

*p < .0 5.

400 Journal of Family Issues 44(2)

contrast, several grandfamily characteristics were associated with sibling- related stress in older children but not in younger children. Overall, positive parenting appears important for both younger and older children, grandparent social support appears important for younger children, and grandparent mental health appears important for older children.

Discussion

In this study we found that children being reared by their grandparents in rural Appalachia exhibited high levels of depression and anxiety symptoms. Children also reported significantly more sibling-related stress than grandparent-related stress. Grandparent parenting stress was positively as- sociated with child depressive symptoms, while grandparent positive par- enting was negatively associated with child depressive symptoms. More children in the household, greater grandparent depression and having a single grandparent were associated with greater child anxiety. Although grandchild adjustment and grandfamily characteristics did not differ across child sex or age, the associations between grandfamily characteristics and child inter- nalizing symptoms differed between boys and girls, and between older and younger children.

Children in our sample experienced high levels of stress, depression and anxiety symptoms. It is interesting that sibling-related stress was significantly higher than grandparent-related stress. Although sibling conflict is to some extent normative, it may be especially distressing in a culture in which family ties and loyalty are emphasized. To our knowledge, this is the first study to examine aspects of the sibling relationship among grandfamilies, and addi- tional research is needed. It is possible that sibling conflict may be exacerbated in families where there are limited resources and competition for grandparent attention (Salmon & Hehman, 2014). We also note that the percent of children with potentially clinical levels of depression is substantially higher than the rates seen in other high-risk samples, such as children and adolescents who have been exposed to trauma (24.7%) and a much larger sample of children involved in the US child welfare system (17% of girls and 7% of boys; Orton, Riggs, & Libby, 2009). Accurate estimates of prevalence rates for a population require large sample sizes, and therefore additional research is needed before concluding that children being reared by grandparents in rural Appalachia are at 2–3 times the risk for internalizing problems than other high risk samples. However, the findings do indicate that the current sample is experiencing high rates of distress.

Although only a preliminary study, the pattern of associations can be used to construct a possible model of child internalizing symptoms in grand- families. In this model, a constellation of risk factors may create an envi- ronment conducive to the experience of children’s stress, anxiety, and

Keller et al. 401

depression. Most distal to the child, grandparents may contend with frailty due to older age, lack of cognitive resources for dealing with their situation and high financial strain due to low education, and issues obtaining formal custody of their grandchildren. Each of these factors was associated with internalizing symptoms for at least some of the children in the study. Such distal risk factors may result in increases in grandparent parenting stress and declines in grandparent mental health, variables also associated with grandchild inter- nalizing symptoms in the present study. In turn, stress and mental illness may impair the parenting skills of grandparents. Although the current study only examined positive parenting, results indicated that this variable was con- sistently associated with lower internalizing symptoms and conversely that less positive parenting would be related to higher internalizing symptoms. Finally, the possible causal chain from distal risk factors to grandparent stress and mental illness to grandparenting to child internalizing symptoms may be moderated by protective factors. Grandparent religiosity and social support were related to lower grandchild internalizing symptoms, and therefore may ameliorate the possible negative effects of the risk factors that can occur in this population.

Although the goal of the current study was to describe risk and protective factors in the context of grandparents rearing grandchildren in rural Appa- lachia, it is worth considering whether the provisional model described above may also apply to other at-risk family types. For example, there are likely to be similarities and differences to models designed for traditional two-parent biological families who are also economically disadvantaged. Similarities include how financial strain and low cognitive resources from poor education may contribute to parenting stress, which in turn reduces positive parenting and may increase hostile parenting. Such family processes are likely to put children in two-parent biological families at risk for internalizing symptoms as well. Such a model has been proposed for economically disadvantaged families and is well-supported (Masarik & Conger, 2017). On the other hand, traditional two-parent biological families are less likely to face issues of physical frailty, custodial issues, and the loss of the parent generation, while at the same time potentially having the disadvantage of less social support (in Appalachia) and life experience.

It is important to note that associations differed based on child sex and age. For boys, many associations involved grandfamily demographics, such as grandparent age, education, number of children, financial status, and whether grandparents had custody of them. In terms of the potential model, this would suggest that there are direct associations between these risk factors and boys’ internalizing symptoms rather than indirect associations through the other variables, or that other possible intervening variables need to be identified. For girls, many associations involved grandparent mental health, depression, and parenting stress, supporting the possible model. Interestingly, there has been

402 Journal of Family Issues 44(2)

very little research on sex differences in child adjustment in grandfamilies. One study found no sex differences in associations between grandfamily functioning and child adjustment in a national sample (Smith, Palmieri, Hancock, & Richardson, 2008). However, research more broadly does re- port gender differences in the impact of family stress on child adjustment, with girls often experiencing greater impacts (Goodman, Rouse, Connell, Broth, Hall, & Heyward, 2011). Girls are often socialized to value and take re- sponsibility for family harmony (Leman & Tenenbaum, 2011); this may especially be the case in rural Appalachia (Latimer & Oberhauser, 2004).

There were also interesting age differences in associations. These dif- ferences concerned the role of protective factors. Grandparent religiosity was related to greater stress among younger children but was unrelated to in- ternalizing symptoms for older children. In addition, grandparent social support was associated with lower internalizing symptoms for younger children, but no associations were observed for older children. Grandparent social support may be less related to outcomes among adolescents because adolescents are building their own social support networks with peers (Shulman & Scharf, 2000). It is unclear why grandparent religiosity would be positively associated with internalizing symptoms for younger children; these associations were not found for older children and negative associations were found for boys. As negative associations between religiosity and mental health problems are typically observed (Sawatzky, Ratner, & Chiu, 2005), additional research is needed.

Social service and other programs that capitalize on these strengths may be of benefit to grandfamilies. Residents of rural Appalachia have limited access to mental health care services (Hendryx, 2008). They may also be wary of outsiders who enter the community with limited understanding of their culture (Weller, 2013). However, Appalachians place a high value on family well- being and are often willing to expend great effort to care for their children (Manoogian, Jurich, Sano, & Ko, 2015). The frequent custodial role that grandparents take in rural Appalachia is a testament to the strength and re- silience of family ties. Culturally adapted and community based interventions for Appalachian grandfamilies, therefore, may be successful. For example, family strengthening programs that have been adapted to Appalachian culture result in higher retention in the program and significantly improve parenting and child outcomes (Marek, Brock, & Sullivan, 2006).

Findings should be interpreted in light of study limitations. First, the current study has a small sample size. Small sample sizes can lead to inflated Type I and Type II error rates, and additional research with larger samples is needed. With a larger sample size, statistical interactions between grandfamily functioning and child sex or age can be directly tested. Our approach of splitting the sample into two groups is only suggestive of interactions, and therefore direct tests of interactions should be done in future research. Second,

Keller et al. 403

the current study has a cross-sectional design, which prevents inferences about causality. We speculate that grandfamily challenges (and strengths) exert a negative (and positive) effect on child internalizing symptoms. However, findings are also consistent with the possibility that children’s mental health problems cause increased stress, mental illness, and physical health problems in grandparents. Previous research has documented that grandparents raising grandchildren with physical, emotional, or behavioral problems experienced more distress than grandparents raising grandchildren without these problems (Hayslip & Shore, 2000). Therefore, future research with a longitudinal design will be able to clarify the likely bidirectional nature of associations. Despite these limitations, the current study improves understanding of the strengths and limitations of grandfamilies in rural Appalachia, adds to the broader research on grandfamilies, and provides some insight into a possible model for the development of grandchild internalizing symptoms in this context.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by a grant awarded to the third author from the Retirement Research Foundation, an Igniting Research Collaborations grant awarded to the first and third authors by the University of Kentucky, and a small grant awarded to the first author from the University of Kentucky Center for Clinical and Translational Science (CCTS). The CCTS is funded by the National Center for Advancing Translational Sciences, National Institutes of Health, through grant number UL1TR001998. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Retirement Research Foundation or the NIH.

ORCID iD

Peggy S. Keller  https://orcid.org/0000-0002-3050-5670

References

Administration for Children and Families (2014). Child maltreatment 2014. Retrieved from http://www.acf.hhs.gov/programs/cb/research-data-technology/statistics- research/child-maltreatment.

Berry, J. O., & Jones, W. H. (1995). The parental stress scale: Initial psychometric evidence. Journal of Social and Personal Relationships, 12(3), 463–472.

404 Journal of Family Issues 44(2)

Brunello, G., & Rocco, L. (2019). Grandparents in the blues. The effect of childcare on grandparents’ depression. Review of Economics of the Household, 17(2), 587–613.

Chorpita, B. F., Tracey, S. A., Brown, T. A., Collica, T. J., & Barlow, D. H. (1997). Assessment of worry in children and adolescents: An adaptation of the Penn State Worry Questionnaire. Behaviour Research and Therapy, 35(6), 569–581.

Connor-Smith, J. K., Compas, B. E., Wadsworth, M. E., Thomsen, A. H., & Saltzman, H. (2000). Responses to stress in adolescence: Measurement of coping and in- voluntary stress responses. Journal of Consulting and Clinical Psychology, 68(6), 976–992.

DeFrain, J. (2014). Strong families around the world: Strengths-based research and perspectives. London: Routledge.

Downie, J. M., Hay, D. A., Horner, B. J., Wichmann, H., & Hislop, A. L. (2010). Children living with their grandparents: Resilience and wellbeing. International Journal of Social Welfare, 19(1), 8–22.

Dunfee, M. N., Brown, R. L., & Schoenberg, N. E. (2021). Perspectives on religious and spiritual coping among rural grandparent caregivers. Journal of Family Is- sues, 42(7), 1495–1515.

Fendrich, M., Weissman, M. M., & Warner, V. (1990). Screening for depressive disorder in children and adolescents: Validating the center for epidemiologic studies depression scale for children. American Journal of Epidemiology, 131(3), 538–551.

Glaser, K., Stuchbury, R., Price, D., Di Gessa, G., Ribe, E., & Tinker, A. (2018). Trends in the prevalence of grandparents living with grandchild (ren) in selected European countries and the United States. European Journal of Ageing, 15(3), 237–250.

Goodman, S. H., Rouse, M. H., Connell, A. M., Broth, M. R., Hall, C. M., & Heyward, D. (2011). Maternal depression and child psychopathology: a meta-analytic review. Clinical child and family psychology review, 14(1), 1–27.

Goulette, N. W., Evans, S. Z., & King, D. (2016). Exploring the behavior of juveniles and young adults raised by custodial grandmothers. Children and Youth Services Review, 70, 349–356.

Hayslip, B., Jr., Fruhauf, C. A., & Fish, J. (2021). Should I do this? Factors influencing the decision to raise grandchildren among custodial grandparents. The Geron- tologist, 61(5), 735–745.

Hayslip, B., & Shore, R. J. (2000). Custodial grandparenting and mental health services. Journal of Mental Health and Aging, 6(4), 367–384.

Hendryx, M. (2008). Mental health professional shortage areas in rural Appalachia. The Journal of Rural Health, 24(2), 179–182.

Jenkins-Howard, S. B., Stephenson, L., & Mains, M. (2013). Cooperative extension nutrition education program: Outreach to Southeast Kentucky families in poverty. PRISM: A Journal of Regional Engagement, 2(2), 117–132.

Kelley, S. J., Whitley, D. M., & Campos, P. E. (2011). Behavior problems in children raised by grandmothers: The role of caregiver distress, family resources, and the home environment. Children and Youth Services Review, 33(11), 2138–2145.

Keller et al. 405

Lamers, F., Jonkers, C. C. M., Bosma, H., Penninx, B. W. J. H., Knottnerus, J. A., & van Eijk, J. T. M. (2008). Summed score of the Patient Health Questionnaire-9 was a reliable and valid method for depression screening in chronically ill elderly patients. Journal of clinical epidemiology, 61(7), 679–687.

Lancki, K., Roberto, K., Savla, J., Borowski, S., Harris, A., Vipperman, A., & Hoyt, E. (2018). For the love of land and people: Cultural reasons for providing care to family members with dementia. Innovation in Aging, 2(Suppl 1), 292.

Latimer, M., &Oberhauser, A. (2004). Exploring gender and economic development in Appalachia. Journal of Appalachian Studies, 10, 269–291.

Leman, P. J., & Tenenbaum, H. R. (2011). Practising gender: Children’s relationships and the development of gendered behaviour and beliefs. British Journal of Developmental Psychology, 29(2), 153–157.

Li, Y., Cui, N., Kok, H. T., Deatrick, J., & Liu, J. (2019). The relationship between parenting styles practiced by grandparents and children’s emotional and be- havioral problems. Journal of Child and Family Studies, 28(7), 1899–1913.

Manoogian, M. M., Jurich, J., Sano, Y., & Ko, J.-L. (2015). “My kids are more important than money” parenting expectations and commitment among Appa- lachian low-income mothers. Journal of Family Issues, 36(3), 326–350.

Marek, L. I., Brock, D.-J. P., & Sullivan, R. (2006). Cultural adaptations to a family life skills program: Implementation in rural Appalachia. The Journal of Primary Prevention, 27(2), 113–133.

Martin, A., Rief, W., Klaiberg, A., & Braehler, E. (2006). Validity of the brief patent health questionnaire mood scale (PHQ-9) in the general population. General Hospital Psychiatry, 28, 71–77.

Masarik, A. S., & Conger, R. D. (2017). Stress and child development: A review of the family stress model. Current opinion in psychology, 13, 85–90.

Nonnemaker, J. M., McNeely, C. A., & Blum, R. W. (2003). Public and private domains of religiosity and adolescent health risk behaviors: Evidence from the National Longitudinal Study of Adolescent Health. Social Science & Medicine, 57(11), 2049–2054.

Orton, H. D., Riggs, P. D., & Libby, A. M. (2009). Prevalence and characteristics of depression and substance use in a US child welfare sample. Children and Youth Services Review, 31(6), 649–653.

Park, M., Unützer, J., & Grembowski, D. (2014). Ethnic and gender variations in the associations between family cohesion, family conflict, and depression in older Asian and Latino adults. Journal of Immigrant and Minority Health, 16(6), 1103–1110.

Pilkauskas, N. V., & Dunifon, R. E. (2016). Understanding grandfamilies: Charac- teristics of grandparents, nonresident parents, and children. Journal of Marriage and Family, 78(3), 623–633.

Pittman, L. D., & Boswell, M. K. (2007). The role of grandmothers in the lives of preschoolers growing up in urban poverty. Applied Developmental Science, 11(1), 20–42.

406 Journal of Family Issues 44(2)

Pollard, K., & Jacobsen, L. A. (2014). The Appalachian region: A data overview from the 2008-2012 American community survey. Washington, DC: Appalachian Regional Commission.

Rapp, AM, Lau, A, & Chavira, DA (2016). Differential associations between social anxiety disorder, family cohesion, and suicidality across racial/ethnic groups: Findings from the National Comorbidity Survey-Adolescent (NCS-A). Journal of Anxiety Disorders, 48, 13–21.

Roberson, P. N., Cortez, G., Trull, L. H., & Lenger, K. (2020). In their ownwords: How Opioids have impacted the lives of“ everyday” people living in Appalachia. Journal of Appalachian Health, 2(4), 26–36.

Rubin, D. M., Downes, K. J., O’reilly, A. L. R., Mekonnen, R., Luan, X., & Localio, R. (2008). Impact of kinship care on behavioral well-being for children in out- of-home care. Archives of Pediatrics & Adolescent Medicine, 162(6), 550–556.

Salmon, C. A., & Hehman, J. A. (2014). The evolutionary psychology of sibling conflict and siblicide. In The evolution of violence (pp. 137–157). New York, NY: Springer.

Sawatzky, R., Ratner, P. A., & Chiu, L. (2005). A meta-analysis of the relationship between spirituality and quality of life. Social Indicators Research, 72(2), 153–188.

Schaefer, E. S. (1965). A configurational analysis of children’s reports of parent behavior. Journal of Consulting Psychology, 29(6), 552–557.

Schalkoff, C. A., Lancaster, K. E., Gaynes, B. N., Wang, V., Pence, B. W., Miller, W. C., & Go, V. F. (2020). The opioid and related drug epidemics in rural Appalachia: A systematic review of populations affected, risk factors, and in- fectious diseases. Substance abuse, 41(1), 35–69.

Schoenberg, N. E., Hatcher, J., & Dignan, M. B. (2008). Appalachian women’s perceptions of their community’s health threats. The Journal of Rural Health, 24(1), 75–83.

Sherbourne, C. D., & Stewart, A. L. (1991). The MOS social support survey. Social Science & Medicine, 32, 705–714.

Shulman, S., & Scharf, M. (2000). Adolescent romantic behaviors and perceptions: Age-and gender-related differences, and links with family and peer relationships. Journal of Research on Adolescence, 10(1), 99–118.

Smith, G. C., & Hancock, G. R. (2010). Custodial grandmother-grandfather dyads: Pathways among marital distress, grandparent dysphoria, parenting practice, and grandchild adjustment. Family relations, 59(1), 45–59.

Smith, G. C., Palmieri, P. A., Hancock, G. R., & Richardson, R. A. (2008). Custodial grandmothers’ psychological distress, dysfunctional parenting, and grandchildren’s adjustment. The International Journal of Aging and Human Development, 67(4), 327–357.

Tang, W., Wang, G., Hu, T., Dai, Q., Xu, J., Yang, Y., & Xu, J. (2018). Mental health and psychosocial problems among Chinese left-behind children: A cross- sectional comparative study. Journal of affective disorders, 241, 133–141.

Keller et al. 407

Vibhakar, V., Allen, L. R., Gee, B., &Meiser-Stedman, R. (2019). A systematic review and meta-analysis on the prevalence of depression in children and adolescents after exposure to trauma. Journal of affective disorders, 255, 77-89.

Waldrop, D. P., & Weber, J. A. (2001). From grandparent to caregiver: The stress and satisfaction of raising grandchildren. Families in Society: The Journal of Con- temporary Social Services, 82(5), 461–472.

Ware, J. E., Kosinski, M., & Keller, S. D. (1994). SF-36 physical and mental health summary scales: A user’s manual. Boston, MA: The Health Institute, New England Medical Center.

Weller, J. E. (2013). Yesterday’s people: Life in contemporary Appalachia. Lexington, KY: University Press of Kentucky.

Whitehead, BR & Bergeman, CS (2020). Daily religious coping buffers the stress– affect relationship and benefits overall metabolic health in older adults. Psy- chology of Religion and Spirituality, 12(4), 393–399.

White, J., Shelton, K. H., & Elgar, F. J. (2014). Prospective associations between the family environment, family cohesion, and psychiatric symptoms among ado- lescent girls. Child Psychiatry & Human Development, 45(5), 544–554.

Xu, Y., Wu, Q., Levkoff, S. E., & Jedwab, M. (2020). Material hardship and parenting stress among grandparent kinship providers during the COVID-19 pandemic: The mediating role of grandparents’ mental health. Child Abuse & Neglect, 110(Pt 2), 104700.

Yoo, J., & Russell, D. W. (2020). Caring for grandchildren and grandparents’ physical and mental health changes. Journal of Child and Family Studies, 29(3), 845–854.

408 Journal of Family Issues 44(2)