Aging & Mental Health, 2015 Vol. 19, No. 4, 315 324, http://dx.doi.org/10.1080/13607863.2014.938606
The pathway to grandparenting stress: trauma, relational conflict, and emotional well-being Ginny Spranga,b*, Moon Choic,d, Jessica G. Eslingera and Adrienne L. Whitt-Woosleya a
Center on Trauma and Children, University of Kentucky, Lexington, KY, USA; bDepartment of Psychiatry, College of Medicine, University of Kentucky, Lexington,KY, USA; cCollege of Social Work, University of Kentucky, Lexington, KY, USA; dSanders-Brown Center on Aging, University of Kentucky, Lexington, KY, USA (Received 14 March 2014; accepted 19 June 2014) Objectives: This paper examines the mediating effect of child grandparent conflict on the relationship between child trauma exposure and grandparenting stress. Methods: Data was collected from a sample of custodial grandparents who participated in kinship care or relative caregiving programs (n D 251). Grandparenting stress was measured with Parenting Stress Scale (Berry & Jones, 1995) modified for grandparents. A series of regression models and structural equation models (SEM) were used to test the relationship between the number of different types of child trauma exposures and grandparenting stress, and to examine the mediating effect of child grandparent conflicts on the relationship. Results: Almost three-fourths (72%) of children had experienced at least one type of traumatic exposure. The SEM model shows that child’s trauma exposure indirectly affected grandparenting stress, mediated by child grandparenting conflicts though no direct path between the child’s trauma exposure variable and grandparenting stress was found. A higher level of child grandparent conflicts was also associated with a lower level of emotional well-being among custodial grandparents. Conclusion: Based on these findings, recommendations are made about how to tailor a trauma-informed approach to the needs of custodial grandparents. Keywords: stress/burden; coping; mental health
Introduction Over the past two decades, the number of grandparents serving as primary caregivers for their grandchildren has steadily increased. According to the US Census, in 2010 there were 2.7 million grandparents who were responsible for the basic needs of one or more grandchildren, with a quarter of these caregivers reporting a personal disability, and over 20% with incomes at or below the poverty level (US Census Bureau, 2012). Projected demographic trajectories predict this trend will continue, necessitating the development of programs and services to address the unique needs of these modern day families. The practice of extended family members participating in the care and raising of children is an American tradition that transcends generations, a practice that was historically precipitated by military deployment, or the death or illness of a parent, conditions that persist today (Bunch, Eastman, & Moore, 2007; Poindexter, 2007). However, the reasons for the current trend towards grandparent-headed households are a reflection of the escalation of more contemporary problems. Caregiver substance abuse, child abuse and neglect, intimate partner violence, and parental incarceration all contribute to the need for grandparent led care (Crewe, 2006; Dellman-Jenkins, Blankemeyer, & Olesh, 2002; Goodman & Rao, 2007; Hirshorn, VanMeter, & Brown, 2000; Kropf & Robinson, 2004). In this context, custodial grandparents are forced to cope with a constellation of personal reactions (e.g. loss, disappointment, anxiety) to their changing role as
*Corresponding author. Email: [email protected] Ó 2014 Taylor & Francis
primary caregiver, as well as challenges parenting children who have faced considerable trauma exposure, and who subsequently may be physiologically and behaviorally dysregulated (Hayslip & Kaminski, 2005; Miltenberger, Hayslip, Harris, & Kaminski, 2003 2004). In fact, Strong, Bean, and Feinauer (2010) noted that the trauma experience influences both grandparents and grandchildren developmentally in terms of their physical and emotional health. Given the nature of the experiences that led to relative placement, investigation of the child grandparent experience from a trauma perspective provides some insight into the child, adult and relational factors that may influence grandparenting stress.
Grandparents raising grandchildren As the number of grandparents raising grandchildren has increased, efforts have focused on understanding the special needs of grandfamilies. Research to date has examined the well-being of grandparents (Baker & Silverstein, 2008a; Butler & Zakari, 2005; Gerard, Landry-Meyer, & Roe, 2006; Hayslip & Glover, 2008; Leder, Grinstead, & Torres, 2007; Neely-Barnes, Graff, & Washington, 2010), health behaviors of grandparents (Baker & Silverstein, 2008b), the role of grandparents (Landry-Meyer & Newman, 2004), policies and laws affecting custodial grandparents (Bratteli, Bjelde, & Pigatti, 2008; Cox, 2009; Robinson-Dooley & Kropf, 2006; Van Etten & Gautam, 2012), services (Campbell, Carthron, Miles, & Brown,
316
G. Sprang et al.
2012; Collins, 2011; Cox, 2008; Kelley, Whitley, & Campos, 2010), the grandparent grandchild relationship (Dolbin-MacNab & Keiley, 2006; Poehlmann et al., 2008), the health and achievement of grandchildren (Smith & Palmieri, 2007), and ethnic and cultural implications for grandfamilies (Carr, Hayslip, & Gray, 2012); Conway, Jones, & Speakes-Lewis, 2011; Kelch-Oliver, 2008). The literature has been primarily cross-sectional in nature and has focused on establishing an empiricallybased understanding of the needs of grandfamilies. Furthermore, the experiences of grandmothers have been a primary focus, suggesting a gap in our understanding of the needs of parenting grandfathers. Due to the formation of grandfamilies frequently resulting from adverse life circumstances, understanding the needs of these family units from a developmental framework grounded in the trauma and attachment literature is indicated (Strong, Bean, & Feinauer, 2010). A trauma framework A trauma perspective is rooted in the understanding that exposure to a traumatic event (any event that causes intense feelings of fear and anxiety) has the potential to interrupt physiological, emotional, and interpersonal processes leading to poor mental and physical health outcomes across the lifespan (Briere, Kaltman, & Green, 2008; Cloitre et al., 2009; Cook et al., 2005; Edwards, Holden, Felitti, & Anda, 2003; Felitti & Anda, 2009). The development of adverse symptoms following traumatic exposure is well documented and can include symptoms of post-traumatic stress (Alisic, Jongmans, van Wesel, & Kleber, 2011, Crusto et al., 2010; Feldman & Vengrober, 2011), depression (Feldman & Vengrober, 2011; Greeson et al., 2011), anxiety (Greeson et al., 2011; Kaplow, Dodge, Amaya-Jackson, & Saxe, 2005), aggression (Ozcol, Zucker, & Spinazzola, 2011), and interpersonal problems (Feldman & Vengrober, 2011; Kim & Cicchetti, 2003). These reactions occur when a child’s internal and external resources are inadequate to cope with perceived threats to their safety. Understanding the relationship between post-trauma distress experienced by the child and the subsequent impact of the grandparent and the grandparent child dyad necessitates the utilization of a transactional model (Sameroff, 2005, 2009). A transactional model specifies that child development occurs as a result of reciprocal interactions between neurobiology, interpersonal relationships, and environmental factors. A traumatic event can interrupt a child’s progression through normative stages of development, potentially leading to traumatic stress-related problems. Studies have indicated that trauma-exposed children often have histories of two or more trauma experiences (Crusto et al., 2010; Finkelhor, Ormrod, Turner, & Hamby, 2005; Greeson et al., 2011). Previous research has found a positive relationship between the number of traumatic experiences and higher levels of emotional and behavioral symptoms (Cloitre et al., 2009; Crusto et al., 2010; Greeson et al., 2011; Kisiel, Fehrenbach, Small, & Lyons, 2009).
The type and frequency of trauma exposure have been found to influence emotional and behavioral symptoms in children (Lau et al., 2005; Luthra et al., 2009). Interpersonal trauma, such as witnessing of domestic violence, physical abuse, and sexual abuse, has been found to predict a diagnosis of post-traumatic stress disorder (PTSD) (Luthra et al., 2009). Furthermore, classifications of abuse and neglect that aim to capture the severity, frequency, and hierarchical nature of such exposures have been found to predict emotional and behavioral outcomes for young children (Lau et al., 2005). Children living with their grandparents have been found to display higher levels of emotional problems when compared to the general population (Smith & Palmieri, 2007), a finding that encourages clarification of the relationship between potential trauma exposure and emotional and relational outcomes for these children. Grandparenting stress Elevated levels of parenting stress have been identified in parenting grandparents. Ross and Aday (2006) found that 92% of custodial grandparents identified clinically significant levels of stress as measured by the Parenting Stress Index (PSI/SF; Abidin, 1995). Other studies have found grandparenting stress to be associated with financial strain (Butler & Zakari, 2005; Minkler, 2005), physical and emotional health of the grandparent (Gerard et al., 2006; Leder et al., 2007), changes in and/or the availability of social support (Butler & Zakari, 2005; Leder et al., 2007), and a grandchild’s emotional and behavior problems (Smith & Palmieri, 2007). Studies have suggested a relationship between child emotional and behavioral problems, parenting stress, and emotional closeness within the grandchild grandparent relationship. A study examining grandparent stress using the PSI (Abidin, 1995) found clinical elevations on both the Parent Child Dysfunctional Interaction and Difficult Child subscales suggesting that difficulties within the child caregiver relationship may influence levels of parenting stress (Ross & Aday, 2006). Lower levels of relational closeness between child and parenting grandparent has been found for grandparents, who identified their grandchildren as having emotional and behavioral problems (Dolbin-MacNab & Keiley, 2006). Furthermore, higher levels of conflict between grandchild and parenting grandparent have been found to be related to poor grandparent health outcomes (Goodman, Tan, Philip, Ernandes, & Silverstein, 2008). Within the general trauma literature, the quality of the child-caregiver relationship has been found to be a protective factor for children following traumatic exposure helping to decrease levels of distress experienced by the child (Becker-Weidman, 2006; Lieberman, 2004). Changes in family roles have also been linked to grandparenting stress and well-being. Landry-Meyer and Newman (2004) found that a majority of parenting grandparent respondents identified experiencing role conflict as they shifted from a traditional grandparenting role to that of full-time parent. Increased grand-caregiver
Aging & Mental Health responsibilities have been found to contribute to increased strain within family relationships and increased the perceived levels of parenting stress (Musil et al., 2011). Lower levels of perceived reward have been reported by parenting grandparents compared to non-parenting grandparents and those living in a multi-generational home (Musil et al., 2011). High levels of parenting stress and lower levels of physical and emotional health have been found for parenting grandparents (Leder et al., 2007).
Study aims This study aims to examine the following hypotheses: (1) The number of different types of trauma exposures experienced by the child will be associated with increases in the level of grandparenting stress in custodial grandparents; (2) Child grandparent conflicts and the emotional wellbeing of the grandparent will be directly associated with the number of different trauma exposures, and (3) these variables will in turn be related to the level of grandparenting stress.
Method Sample A total of 297 custodial grandparents participated in the study. The sampling process consisted of two steps. First, local conferences and events were attended to solicit participation via verbal invitations by study personnel and announcements by conference organizers. Second, program coordinators for the statewide care network from two multi-county regions sent individual correspondence to each member of their program requesting they complete the survey online or by phone. Participants were encouraged to forward the information about the survey to other custodial grandparents. Due to the use of a snowball sampling strategy, an estimated response rate was not available. Additionally, no official list of the universe of custodial grandparents exists, but it was the aim of this study to encourage the participation of as many grandparents raising grandchildren as possible. All participants solicited directly were given the opportunity to complete the survey either in written form, online through Survey Monkey or via live interview. Ten participants elected to complete the survey by phone, 30 participants completed the survey online and the remaining 257 participants completed a paper version of the survey. Sample participants were 83.9% female, and 78.7% white. Both rural and more urban areas of the state were purposely sampled in order to attempt to obtain a racially diverse sample (since most minority groups in the state reside in urban areas), and subsequently the racial distribution in the sample mirrors population statistics in the state. Approximately 43.7% reported being married. The average grandparent age was 59.55 (SD D 9.50) years, and there was a wide range of ages observed in the sample from 34 to 81 years. The average years of education reported was 12.1 (SD D 3.33). Almost half of the sample
317
identified residing in metropolitan areas (Beale codes 1 3), 15.8% in non-metro areas with urban influence (Beale codes 4 and 6), 15.1% in non-metro areas with no urban influence (Beale Codes 5and 7), 6.3% lived in areas that were completely rural with urban influence (Beale Code 8), and 14% lived in completely rural areas with no urban influence. The custodial grandparents in this study were parenting an average of two children (SD 1.15), 53.1% were caring for children 12 years of age or older, 55% were raising children 6 11 years of age, and over one-third (37.8%) had assumed caregiving responsibility for children of age 5 or younger. The most frequently cited reasons for grandchildren coming into the primary care of a grandparent included substance misuse (46.1%), child welfare involvement (31%), incarceration of a parent (23.5%), death of a parent, financial problems (16.2%), and death of a parent (12.8%). Study participants were offered a $10 incentive for each completed interview or survey, which was estimated to take approximately 45 minutes to complete. A consent form was included with each survey or interview that explained the risks and benefits of participation. Participants were assured that their information would be kept private and confidential, no identifying information was collected, and it was explained that there were minimal risks associated with their participation. All participation was voluntary and it was communicated to the participants that there were no penalties associated with refusal to complete the survey. All protocols were approved by the University of Kentucky Internal Review Board prior to initiation of the study.
Measurement The questionnaire included several standardized instruments widely accepted in the literature to assess caregiver stress, the caregiving relationship, health and individual functioning. General questions about grandparent’s parenting stress and health, child’s exposure to traumatic events, and basic demographic information were also included. A description of the measures and additional questions included in the survey is provided below.
Caregiver factors The Parental Stress Scale was included to measure caregiver stress via assessment of both positive and negative themes associated with parenting, including emotional benefits, self-enrichment, demand on resources and restrictions (Berry & Jones, 1995). Sample items asked respondents to rate whether they strongly disagree, disagree, are undecided, agree or strongly agree with statements such as, ‘I am happy in my role as a parent,’ ‘Having children is a financial burden,’ or ‘I am satisfied as a parent.’ This 18-item, self-report instrument has demonstrated the ability to discriminate between parental reports of clinical and non-clinical samples (Berry & Jones, 1995). Cronbach’s alpha for this scale was .719 for this sample, indicating good internal consistency.
318
G. Sprang et al.
The RAND SF 36-Item Health Survey (SF, short form) was included in the survey to assess caregivers’ experiences of health regarding physical functioning, role limitations due to physical functioning or emotional health, energy/fatigue, emotional well-being, social functioning, pain, general health, and comorbidity (Hays, Sherbourne, & Mazel, 1993). Multiple studies of this measure have found strong psychometric properties of validity and reliability (Brazier et al., 1992; Garatt, Ruta, Abdalla, Buckingham, & Russell, 1993; Vander Zee, Sanderman, Heynik, & de Haes, 1996). Child factors Trauma Exposure. Participants were asked whether any grandchild in their care had been exposed to a traumatic event prior to grandparent assumption of primary care. Options of no, yes, suspected or unknown were offered on this item. If the caregiver positively endorsed this item, then they were asked to check all types of traumatic experience that might apply. Options offered on this item included no trauma exposure, sexual maltreatment, physical maltreatment, emotional maltreatment, neglect, domestic violence, war/terrorism/political violence, illness/medical trauma, serious injury or accident, natural disaster, kidnapping, traumatic loss, impaired caregiver, community violence, school violence or other (participant defined category). The numbers of different traumatic events endorsed were summed to provide the frequency of exposure to discrete types of events for each child. Grandparents were then asked if any grandchild in their care had been diagnosed with an anxiety disorder, or traumarelated disorder such as post-traumatic stress disorder, or acute stress disorder. If they positively endorsed these items, respondents were asked to explain which disorder applied. Relationship factors The child grandparent conflicts were assessed with the Child Parent Relationship Scale (CRPS; Pianta, 1992), which measures qualities of the relationship as reported by caregivers through the constructs of conflicts and closeness Only the eight-item conflicts subscale was included in the survey out of concern for respondent burden. Questions included asking about whether there are struggles in the relationship, degree of comfort with physical affection and impact of relationship on caregiver’s energy level. Higher scores on this measure indicated increased levels of grandparent child conflicts. Cronbach’s alpha for the CPRS conflicts subscale was .909 for this sample, indicating excellent internal consistency. Demographic factors Questions were included to elicit descriptive information regarding the grandparents’ socio-demographic characteristics. Demographic questions included assessment of age, gender, race, years of education completed, and
marital status (married, divorced, separated, widowed, never married). Grandparents were queried about the number of children living in their home in broad age categories (0 5, 6 11, 12 18). Statistical analysis First, to examine the differences between grandparents with child exposed to trauma and grandparents with child without any trauma exposure, a comparison of these two groups in terms of their parental stress, child grandparent conflicts, sociodemographic characteristics, and health status was conducted. Statistical significance in differences between the two groups was assessed with chi-square test for categorical variables and t-test for continuous variables. Bivariate analyses were used to test the correlations between key variables, such as the number of different types of trauma, grandparenting stress, child grandparent conflicts, and a grandparent’s emotional well-being. Second, a series of linear regression models were run to examine the mediating effect of child grandparent conflicts between number of different types of trauma (i.e. predictor) and grandparenting stress (i.e. outcome). Finally, a full structural model was estimated to simultaneously examine the relationship between a child’s trauma exposure and child grandparent conflicts, and between child grandparent conflicts and grandparenting stress. Results On average, grandparents in this sample were parenting two children (SD 1.15, range 1 7), 53% were caring for those 12 years of age or older, 55% had children 6 11 years old in their care, and 37.8% of the sample had assumed custody of children 5 years of age or under. About 72% of participants reported that their grandchild had been exposed to trauma. Over half of participants (53.4%) reported that their grandchild had experienced between one and three types of trauma, with a maximum of 12 discrete events identified (Mean D 2.0, SD D 2.0). The most prevalent types of trauma reported by grandparents were emotional/psychological abuse or maltreatment (30%), neglect (27%), impaired caregivers such as exposure to caregiver alcohol or drug use (27%), domestic violence (24%), physical abuse (21%), traumatic loss such as, sudden death or separation (21%), and sexual maltreatment/abuse/assault (18%). However, only 16.3% (41 out of 251 grandparents) reported that their child had been diagnosed with a trauma-related disorder such as posttraumatic stress disorder, acute stress disorder, or anxiety disorder. Table 1 displays the comparison between grandparents with child without any trauma exposure and grandparents with child exposed to trauma, in terms of their grandparenting stress, sociodemographic characteristics, health status, and conflicts with grandchild. Grandparents with a child who had been exposed to trauma reported a higher level of parental stress compared to those with child who had not been exposed to trauma (Parental Stress Scale
Aging & Mental Health
319
Table 1. Characteristics of study participants by their child’s trauma exposure. Grandparent with child without any trauma exposure (n D 70)
Grandparent with child exposed to trauma (n D 181)
55.6 (9.0)
53.1 (9.3)
56.5 (8.8)
.008
59.9 (9.6) 208 (83.9) 196 (78.7) 108 (43.7) 12.1 (3.4) 86 (36.1)
58.2 (8.3) 59 (85.5) 51 (73.9) 29 (42.6) 11.6 (3.0) 24 (36.4)
60.5 (9.9) 149 (83.2) 145 (80.6) 79 (44.1) 12.3 (3.5) 62 (36.0)
.074 .664 .252 .833 .119 .964
59.8 (31.3) 49.3 (45.6) 65.9 (45.3) 45.9 (23.0) 66.0 (21.0) 48.0 (14.5) 42.4 (27.7) 54.3 (16.9) 2.3 (1.5)
56.9 (33.1) 54.5 (44.9) 73.1 (41.1) 47.5 (24.4) 72.1 (19.7) 46.4 (14.1) 42.8 (27.4) 57.0 (17.5) 2.2 (1.3)
61.0 (30.6) 47.3 (45.8) 63.1 (46.5) 45.3 (22.5) 63.7 (21.1) 48.6 (14.6) 42.3 (27.9) 53.3 (16.3) 2.4 (1.6)
.360 .273 .104 .495 .010 .279 .886 .117 .404
21.6 (9.3)
17.3 (8.5)
23.3 (9.1)
All (n D 251) Parental stress Parental stress scale score, M (SD) Grandparent’s sociodemographic characteristics Age, M (SD) C Women, n (%)C White, n (%) Married, n (%)C Years of education, M (SD) C Eastern Kentucky residence, n (%)C Grandparent’s health status Physical functioning, M (SD) C Role limitations due to physical functioning, M (SD) C Role limitations due to emotional health, M (SD) C Energy/fatigue, M (SD) C Emotional well-being, M (SD) Social functioning, M (SD) C Pain, M (SD) C General health, M (SD) C Comorbidity, M (SD) C Child grandparent conflicts Relationship scale, M (SD)
P
< .001
C
These variables had 2 15 cases with missing values. p-value from chi-squared tests for categorical variables and t-tests for continuous variables.
Score: 56.5 and 53.1, respectively; p D .008). There was no statistically significant difference between the two groups in terms of sociodemographic characteristics. However, the difference in grandparent age approached significance. Grandparents with child who had been exposed to trauma were slightly older than those with child who had not been exposed to trauma (58.2 and 60.5 years old, respectively; p D .074). About grandparent’s health status, the only significant difference was found in emotional well-being. Grandparents with a child who had been exposed to trauma reported a lower level of emotional well-being compared to those with a child with no exposure (63.7 and 72.1, respectively; p D .010). Grandparents with child in the trauma exposure group also reported a higher level of child grandparent conflict compared to those with child who had not been exposed to trauma (23.3 and 17.3, respectively; p
The pathway to grandparenting stress: trauma, relational conflict, and emotional well-being Ginny Spranga,b*, Moon Choic,d, Jessica G. Eslingera and Adrienne L. Whitt-Woosleya a
Center on Trauma and Children, University of Kentucky, Lexington, KY, USA; bDepartment of Psychiatry, College of Medicine, University of Kentucky, Lexington,KY, USA; cCollege of Social Work, University of Kentucky, Lexington, KY, USA; dSanders-Brown Center on Aging, University of Kentucky, Lexington, KY, USA (Received 14 March 2014; accepted 19 June 2014) Objectives: This paper examines the mediating effect of child grandparent conflict on the relationship between child trauma exposure and grandparenting stress. Methods: Data was collected from a sample of custodial grandparents who participated in kinship care or relative caregiving programs (n D 251). Grandparenting stress was measured with Parenting Stress Scale (Berry & Jones, 1995) modified for grandparents. A series of regression models and structural equation models (SEM) were used to test the relationship between the number of different types of child trauma exposures and grandparenting stress, and to examine the mediating effect of child grandparent conflicts on the relationship. Results: Almost three-fourths (72%) of children had experienced at least one type of traumatic exposure. The SEM model shows that child’s trauma exposure indirectly affected grandparenting stress, mediated by child grandparenting conflicts though no direct path between the child’s trauma exposure variable and grandparenting stress was found. A higher level of child grandparent conflicts was also associated with a lower level of emotional well-being among custodial grandparents. Conclusion: Based on these findings, recommendations are made about how to tailor a trauma-informed approach to the needs of custodial grandparents. Keywords: stress/burden; coping; mental health
Introduction Over the past two decades, the number of grandparents serving as primary caregivers for their grandchildren has steadily increased. According to the US Census, in 2010 there were 2.7 million grandparents who were responsible for the basic needs of one or more grandchildren, with a quarter of these caregivers reporting a personal disability, and over 20% with incomes at or below the poverty level (US Census Bureau, 2012). Projected demographic trajectories predict this trend will continue, necessitating the development of programs and services to address the unique needs of these modern day families. The practice of extended family members participating in the care and raising of children is an American tradition that transcends generations, a practice that was historically precipitated by military deployment, or the death or illness of a parent, conditions that persist today (Bunch, Eastman, & Moore, 2007; Poindexter, 2007). However, the reasons for the current trend towards grandparent-headed households are a reflection of the escalation of more contemporary problems. Caregiver substance abuse, child abuse and neglect, intimate partner violence, and parental incarceration all contribute to the need for grandparent led care (Crewe, 2006; Dellman-Jenkins, Blankemeyer, & Olesh, 2002; Goodman & Rao, 2007; Hirshorn, VanMeter, & Brown, 2000; Kropf & Robinson, 2004). In this context, custodial grandparents are forced to cope with a constellation of personal reactions (e.g. loss, disappointment, anxiety) to their changing role as
*Corresponding author. Email: [email protected] Ó 2014 Taylor & Francis
primary caregiver, as well as challenges parenting children who have faced considerable trauma exposure, and who subsequently may be physiologically and behaviorally dysregulated (Hayslip & Kaminski, 2005; Miltenberger, Hayslip, Harris, & Kaminski, 2003 2004). In fact, Strong, Bean, and Feinauer (2010) noted that the trauma experience influences both grandparents and grandchildren developmentally in terms of their physical and emotional health. Given the nature of the experiences that led to relative placement, investigation of the child grandparent experience from a trauma perspective provides some insight into the child, adult and relational factors that may influence grandparenting stress.
Grandparents raising grandchildren As the number of grandparents raising grandchildren has increased, efforts have focused on understanding the special needs of grandfamilies. Research to date has examined the well-being of grandparents (Baker & Silverstein, 2008a; Butler & Zakari, 2005; Gerard, Landry-Meyer, & Roe, 2006; Hayslip & Glover, 2008; Leder, Grinstead, & Torres, 2007; Neely-Barnes, Graff, & Washington, 2010), health behaviors of grandparents (Baker & Silverstein, 2008b), the role of grandparents (Landry-Meyer & Newman, 2004), policies and laws affecting custodial grandparents (Bratteli, Bjelde, & Pigatti, 2008; Cox, 2009; Robinson-Dooley & Kropf, 2006; Van Etten & Gautam, 2012), services (Campbell, Carthron, Miles, & Brown,
316
G. Sprang et al.
2012; Collins, 2011; Cox, 2008; Kelley, Whitley, & Campos, 2010), the grandparent grandchild relationship (Dolbin-MacNab & Keiley, 2006; Poehlmann et al., 2008), the health and achievement of grandchildren (Smith & Palmieri, 2007), and ethnic and cultural implications for grandfamilies (Carr, Hayslip, & Gray, 2012); Conway, Jones, & Speakes-Lewis, 2011; Kelch-Oliver, 2008). The literature has been primarily cross-sectional in nature and has focused on establishing an empiricallybased understanding of the needs of grandfamilies. Furthermore, the experiences of grandmothers have been a primary focus, suggesting a gap in our understanding of the needs of parenting grandfathers. Due to the formation of grandfamilies frequently resulting from adverse life circumstances, understanding the needs of these family units from a developmental framework grounded in the trauma and attachment literature is indicated (Strong, Bean, & Feinauer, 2010). A trauma framework A trauma perspective is rooted in the understanding that exposure to a traumatic event (any event that causes intense feelings of fear and anxiety) has the potential to interrupt physiological, emotional, and interpersonal processes leading to poor mental and physical health outcomes across the lifespan (Briere, Kaltman, & Green, 2008; Cloitre et al., 2009; Cook et al., 2005; Edwards, Holden, Felitti, & Anda, 2003; Felitti & Anda, 2009). The development of adverse symptoms following traumatic exposure is well documented and can include symptoms of post-traumatic stress (Alisic, Jongmans, van Wesel, & Kleber, 2011, Crusto et al., 2010; Feldman & Vengrober, 2011), depression (Feldman & Vengrober, 2011; Greeson et al., 2011), anxiety (Greeson et al., 2011; Kaplow, Dodge, Amaya-Jackson, & Saxe, 2005), aggression (Ozcol, Zucker, & Spinazzola, 2011), and interpersonal problems (Feldman & Vengrober, 2011; Kim & Cicchetti, 2003). These reactions occur when a child’s internal and external resources are inadequate to cope with perceived threats to their safety. Understanding the relationship between post-trauma distress experienced by the child and the subsequent impact of the grandparent and the grandparent child dyad necessitates the utilization of a transactional model (Sameroff, 2005, 2009). A transactional model specifies that child development occurs as a result of reciprocal interactions between neurobiology, interpersonal relationships, and environmental factors. A traumatic event can interrupt a child’s progression through normative stages of development, potentially leading to traumatic stress-related problems. Studies have indicated that trauma-exposed children often have histories of two or more trauma experiences (Crusto et al., 2010; Finkelhor, Ormrod, Turner, & Hamby, 2005; Greeson et al., 2011). Previous research has found a positive relationship between the number of traumatic experiences and higher levels of emotional and behavioral symptoms (Cloitre et al., 2009; Crusto et al., 2010; Greeson et al., 2011; Kisiel, Fehrenbach, Small, & Lyons, 2009).
The type and frequency of trauma exposure have been found to influence emotional and behavioral symptoms in children (Lau et al., 2005; Luthra et al., 2009). Interpersonal trauma, such as witnessing of domestic violence, physical abuse, and sexual abuse, has been found to predict a diagnosis of post-traumatic stress disorder (PTSD) (Luthra et al., 2009). Furthermore, classifications of abuse and neglect that aim to capture the severity, frequency, and hierarchical nature of such exposures have been found to predict emotional and behavioral outcomes for young children (Lau et al., 2005). Children living with their grandparents have been found to display higher levels of emotional problems when compared to the general population (Smith & Palmieri, 2007), a finding that encourages clarification of the relationship between potential trauma exposure and emotional and relational outcomes for these children. Grandparenting stress Elevated levels of parenting stress have been identified in parenting grandparents. Ross and Aday (2006) found that 92% of custodial grandparents identified clinically significant levels of stress as measured by the Parenting Stress Index (PSI/SF; Abidin, 1995). Other studies have found grandparenting stress to be associated with financial strain (Butler & Zakari, 2005; Minkler, 2005), physical and emotional health of the grandparent (Gerard et al., 2006; Leder et al., 2007), changes in and/or the availability of social support (Butler & Zakari, 2005; Leder et al., 2007), and a grandchild’s emotional and behavior problems (Smith & Palmieri, 2007). Studies have suggested a relationship between child emotional and behavioral problems, parenting stress, and emotional closeness within the grandchild grandparent relationship. A study examining grandparent stress using the PSI (Abidin, 1995) found clinical elevations on both the Parent Child Dysfunctional Interaction and Difficult Child subscales suggesting that difficulties within the child caregiver relationship may influence levels of parenting stress (Ross & Aday, 2006). Lower levels of relational closeness between child and parenting grandparent has been found for grandparents, who identified their grandchildren as having emotional and behavioral problems (Dolbin-MacNab & Keiley, 2006). Furthermore, higher levels of conflict between grandchild and parenting grandparent have been found to be related to poor grandparent health outcomes (Goodman, Tan, Philip, Ernandes, & Silverstein, 2008). Within the general trauma literature, the quality of the child-caregiver relationship has been found to be a protective factor for children following traumatic exposure helping to decrease levels of distress experienced by the child (Becker-Weidman, 2006; Lieberman, 2004). Changes in family roles have also been linked to grandparenting stress and well-being. Landry-Meyer and Newman (2004) found that a majority of parenting grandparent respondents identified experiencing role conflict as they shifted from a traditional grandparenting role to that of full-time parent. Increased grand-caregiver
Aging & Mental Health responsibilities have been found to contribute to increased strain within family relationships and increased the perceived levels of parenting stress (Musil et al., 2011). Lower levels of perceived reward have been reported by parenting grandparents compared to non-parenting grandparents and those living in a multi-generational home (Musil et al., 2011). High levels of parenting stress and lower levels of physical and emotional health have been found for parenting grandparents (Leder et al., 2007).
Study aims This study aims to examine the following hypotheses: (1) The number of different types of trauma exposures experienced by the child will be associated with increases in the level of grandparenting stress in custodial grandparents; (2) Child grandparent conflicts and the emotional wellbeing of the grandparent will be directly associated with the number of different trauma exposures, and (3) these variables will in turn be related to the level of grandparenting stress.
Method Sample A total of 297 custodial grandparents participated in the study. The sampling process consisted of two steps. First, local conferences and events were attended to solicit participation via verbal invitations by study personnel and announcements by conference organizers. Second, program coordinators for the statewide care network from two multi-county regions sent individual correspondence to each member of their program requesting they complete the survey online or by phone. Participants were encouraged to forward the information about the survey to other custodial grandparents. Due to the use of a snowball sampling strategy, an estimated response rate was not available. Additionally, no official list of the universe of custodial grandparents exists, but it was the aim of this study to encourage the participation of as many grandparents raising grandchildren as possible. All participants solicited directly were given the opportunity to complete the survey either in written form, online through Survey Monkey or via live interview. Ten participants elected to complete the survey by phone, 30 participants completed the survey online and the remaining 257 participants completed a paper version of the survey. Sample participants were 83.9% female, and 78.7% white. Both rural and more urban areas of the state were purposely sampled in order to attempt to obtain a racially diverse sample (since most minority groups in the state reside in urban areas), and subsequently the racial distribution in the sample mirrors population statistics in the state. Approximately 43.7% reported being married. The average grandparent age was 59.55 (SD D 9.50) years, and there was a wide range of ages observed in the sample from 34 to 81 years. The average years of education reported was 12.1 (SD D 3.33). Almost half of the sample
317
identified residing in metropolitan areas (Beale codes 1 3), 15.8% in non-metro areas with urban influence (Beale codes 4 and 6), 15.1% in non-metro areas with no urban influence (Beale Codes 5and 7), 6.3% lived in areas that were completely rural with urban influence (Beale Code 8), and 14% lived in completely rural areas with no urban influence. The custodial grandparents in this study were parenting an average of two children (SD 1.15), 53.1% were caring for children 12 years of age or older, 55% were raising children 6 11 years of age, and over one-third (37.8%) had assumed caregiving responsibility for children of age 5 or younger. The most frequently cited reasons for grandchildren coming into the primary care of a grandparent included substance misuse (46.1%), child welfare involvement (31%), incarceration of a parent (23.5%), death of a parent, financial problems (16.2%), and death of a parent (12.8%). Study participants were offered a $10 incentive for each completed interview or survey, which was estimated to take approximately 45 minutes to complete. A consent form was included with each survey or interview that explained the risks and benefits of participation. Participants were assured that their information would be kept private and confidential, no identifying information was collected, and it was explained that there were minimal risks associated with their participation. All participation was voluntary and it was communicated to the participants that there were no penalties associated with refusal to complete the survey. All protocols were approved by the University of Kentucky Internal Review Board prior to initiation of the study.
Measurement The questionnaire included several standardized instruments widely accepted in the literature to assess caregiver stress, the caregiving relationship, health and individual functioning. General questions about grandparent’s parenting stress and health, child’s exposure to traumatic events, and basic demographic information were also included. A description of the measures and additional questions included in the survey is provided below.
Caregiver factors The Parental Stress Scale was included to measure caregiver stress via assessment of both positive and negative themes associated with parenting, including emotional benefits, self-enrichment, demand on resources and restrictions (Berry & Jones, 1995). Sample items asked respondents to rate whether they strongly disagree, disagree, are undecided, agree or strongly agree with statements such as, ‘I am happy in my role as a parent,’ ‘Having children is a financial burden,’ or ‘I am satisfied as a parent.’ This 18-item, self-report instrument has demonstrated the ability to discriminate between parental reports of clinical and non-clinical samples (Berry & Jones, 1995). Cronbach’s alpha for this scale was .719 for this sample, indicating good internal consistency.
318
G. Sprang et al.
The RAND SF 36-Item Health Survey (SF, short form) was included in the survey to assess caregivers’ experiences of health regarding physical functioning, role limitations due to physical functioning or emotional health, energy/fatigue, emotional well-being, social functioning, pain, general health, and comorbidity (Hays, Sherbourne, & Mazel, 1993). Multiple studies of this measure have found strong psychometric properties of validity and reliability (Brazier et al., 1992; Garatt, Ruta, Abdalla, Buckingham, & Russell, 1993; Vander Zee, Sanderman, Heynik, & de Haes, 1996). Child factors Trauma Exposure. Participants were asked whether any grandchild in their care had been exposed to a traumatic event prior to grandparent assumption of primary care. Options of no, yes, suspected or unknown were offered on this item. If the caregiver positively endorsed this item, then they were asked to check all types of traumatic experience that might apply. Options offered on this item included no trauma exposure, sexual maltreatment, physical maltreatment, emotional maltreatment, neglect, domestic violence, war/terrorism/political violence, illness/medical trauma, serious injury or accident, natural disaster, kidnapping, traumatic loss, impaired caregiver, community violence, school violence or other (participant defined category). The numbers of different traumatic events endorsed were summed to provide the frequency of exposure to discrete types of events for each child. Grandparents were then asked if any grandchild in their care had been diagnosed with an anxiety disorder, or traumarelated disorder such as post-traumatic stress disorder, or acute stress disorder. If they positively endorsed these items, respondents were asked to explain which disorder applied. Relationship factors The child grandparent conflicts were assessed with the Child Parent Relationship Scale (CRPS; Pianta, 1992), which measures qualities of the relationship as reported by caregivers through the constructs of conflicts and closeness Only the eight-item conflicts subscale was included in the survey out of concern for respondent burden. Questions included asking about whether there are struggles in the relationship, degree of comfort with physical affection and impact of relationship on caregiver’s energy level. Higher scores on this measure indicated increased levels of grandparent child conflicts. Cronbach’s alpha for the CPRS conflicts subscale was .909 for this sample, indicating excellent internal consistency. Demographic factors Questions were included to elicit descriptive information regarding the grandparents’ socio-demographic characteristics. Demographic questions included assessment of age, gender, race, years of education completed, and
marital status (married, divorced, separated, widowed, never married). Grandparents were queried about the number of children living in their home in broad age categories (0 5, 6 11, 12 18). Statistical analysis First, to examine the differences between grandparents with child exposed to trauma and grandparents with child without any trauma exposure, a comparison of these two groups in terms of their parental stress, child grandparent conflicts, sociodemographic characteristics, and health status was conducted. Statistical significance in differences between the two groups was assessed with chi-square test for categorical variables and t-test for continuous variables. Bivariate analyses were used to test the correlations between key variables, such as the number of different types of trauma, grandparenting stress, child grandparent conflicts, and a grandparent’s emotional well-being. Second, a series of linear regression models were run to examine the mediating effect of child grandparent conflicts between number of different types of trauma (i.e. predictor) and grandparenting stress (i.e. outcome). Finally, a full structural model was estimated to simultaneously examine the relationship between a child’s trauma exposure and child grandparent conflicts, and between child grandparent conflicts and grandparenting stress. Results On average, grandparents in this sample were parenting two children (SD 1.15, range 1 7), 53% were caring for those 12 years of age or older, 55% had children 6 11 years old in their care, and 37.8% of the sample had assumed custody of children 5 years of age or under. About 72% of participants reported that their grandchild had been exposed to trauma. Over half of participants (53.4%) reported that their grandchild had experienced between one and three types of trauma, with a maximum of 12 discrete events identified (Mean D 2.0, SD D 2.0). The most prevalent types of trauma reported by grandparents were emotional/psychological abuse or maltreatment (30%), neglect (27%), impaired caregivers such as exposure to caregiver alcohol or drug use (27%), domestic violence (24%), physical abuse (21%), traumatic loss such as, sudden death or separation (21%), and sexual maltreatment/abuse/assault (18%). However, only 16.3% (41 out of 251 grandparents) reported that their child had been diagnosed with a trauma-related disorder such as posttraumatic stress disorder, acute stress disorder, or anxiety disorder. Table 1 displays the comparison between grandparents with child without any trauma exposure and grandparents with child exposed to trauma, in terms of their grandparenting stress, sociodemographic characteristics, health status, and conflicts with grandchild. Grandparents with a child who had been exposed to trauma reported a higher level of parental stress compared to those with child who had not been exposed to trauma (Parental Stress Scale
Aging & Mental Health
319
Table 1. Characteristics of study participants by their child’s trauma exposure. Grandparent with child without any trauma exposure (n D 70)
Grandparent with child exposed to trauma (n D 181)
55.6 (9.0)
53.1 (9.3)
56.5 (8.8)
.008
59.9 (9.6) 208 (83.9) 196 (78.7) 108 (43.7) 12.1 (3.4) 86 (36.1)
58.2 (8.3) 59 (85.5) 51 (73.9) 29 (42.6) 11.6 (3.0) 24 (36.4)
60.5 (9.9) 149 (83.2) 145 (80.6) 79 (44.1) 12.3 (3.5) 62 (36.0)
.074 .664 .252 .833 .119 .964
59.8 (31.3) 49.3 (45.6) 65.9 (45.3) 45.9 (23.0) 66.0 (21.0) 48.0 (14.5) 42.4 (27.7) 54.3 (16.9) 2.3 (1.5)
56.9 (33.1) 54.5 (44.9) 73.1 (41.1) 47.5 (24.4) 72.1 (19.7) 46.4 (14.1) 42.8 (27.4) 57.0 (17.5) 2.2 (1.3)
61.0 (30.6) 47.3 (45.8) 63.1 (46.5) 45.3 (22.5) 63.7 (21.1) 48.6 (14.6) 42.3 (27.9) 53.3 (16.3) 2.4 (1.6)
.360 .273 .104 .495 .010 .279 .886 .117 .404
21.6 (9.3)
17.3 (8.5)
23.3 (9.1)
All (n D 251) Parental stress Parental stress scale score, M (SD) Grandparent’s sociodemographic characteristics Age, M (SD) C Women, n (%)C White, n (%) Married, n (%)C Years of education, M (SD) C Eastern Kentucky residence, n (%)C Grandparent’s health status Physical functioning, M (SD) C Role limitations due to physical functioning, M (SD) C Role limitations due to emotional health, M (SD) C Energy/fatigue, M (SD) C Emotional well-being, M (SD) Social functioning, M (SD) C Pain, M (SD) C General health, M (SD) C Comorbidity, M (SD) C Child grandparent conflicts Relationship scale, M (SD)
P
< .001
C
These variables had 2 15 cases with missing values. p-value from chi-squared tests for categorical variables and t-tests for continuous variables.
Score: 56.5 and 53.1, respectively; p D .008). There was no statistically significant difference between the two groups in terms of sociodemographic characteristics. However, the difference in grandparent age approached significance. Grandparents with child who had been exposed to trauma were slightly older than those with child who had not been exposed to trauma (58.2 and 60.5 years old, respectively; p D .074). About grandparent’s health status, the only significant difference was found in emotional well-being. Grandparents with a child who had been exposed to trauma reported a lower level of emotional well-being compared to those with a child with no exposure (63.7 and 72.1, respectively; p D .010). Grandparents with child in the trauma exposure group also reported a higher level of child grandparent conflict compared to those with child who had not been exposed to trauma (23.3 and 17.3, respectively; p
